Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Biology of Gnathotrichus sulcatus (LeConte 1868) (Col.:Scolytidae) with special emphasis on host colonization… Zanuncio, José Cola 1981

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1981_A1 Z36.pdf [ 6.62MB ]
Metadata
JSON: 831-1.0095129.json
JSON-LD: 831-1.0095129-ld.json
RDF/XML (Pretty): 831-1.0095129-rdf.xml
RDF/JSON: 831-1.0095129-rdf.json
Turtle: 831-1.0095129-turtle.txt
N-Triples: 831-1.0095129-rdf-ntriples.txt
Original Record: 831-1.0095129-source.json
Full Text
831-1.0095129-fulltext.txt
Citation
831-1.0095129.ris

Full Text

Biology of Gnathotrichus sulcatus (LeConte 1868) (Col.:Scolytidae) with Special Emphasis on Host Colonization and Brood Production by Jose Cola Zanuncio For. Eng., Federal University of Vijosa, Brazil M.Sc, University of Sao Paulo, Brazil A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Forestry We accept this thesis as conforming to the required standard The University of British Columbia August, 1981 © JosS Cola Zanuncio In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or p u b l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Fo&GST&'l? The University of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date Q c T O 3 € ; / ^ 1 ) DE-6 (2/71) i ABSTRACT The bionomics of Gnathotrichus sulcatus (LeConte) (Col.: Scolytidae) were studied on eight western hemlock (WH), Tsuga  heterophylla (Raf.) Sarg., and eight Douglas-fir (Df), Pseudotsuga  menziesii (Mirb.) Franco, trees at the UBC Research Forest, Maple Ridge, B.C. during 1978 and 1979. Natural colonization by associated scolytid beetles was also studied on the same trees, which were felled on May 8, 1978. More attacks occurred on downtrees (with branches.) than on logs (no branches), and on WH stumps, trunks (logs and down-trees) than on Df stumps and trunks. WH was more suitable for coloni-zation and brood production in the f i r s t year after f e l l i n g than Df but no differences could be demonstrated between these two hosts in the second year. Whenever possible five fresh attacks were covered with petri dishes on every stump and trunk. Higher quantities of frass, numbers of brood, pupal niches, deeper penetration and longer galleries were recorded on WH than on Df. Relationships between brood, frass produc-tion and length of galleries were established. It was calculated that one hectare of clearcut with 200 WH stumps could produce 714,860 + 417,020 beetles with 13.7 + 3.7 brood per gallery (both with 95% confidence limits), within two years after attack, while the estimated total brood production from the 16 trunk/stump combinations was 107,080 from 16,049 attacks. Pupal niches, length and depth of penetration of galleries were studied on attacks initiated two to seventeen months after f e l l i n g , i i excluding the winter months. For each of the three factors WH showed higher results than Df. Traps set out in Point Roberts, Washington, captured more beetles when baited with WH sapwood than with WH bark or heartwood. More beetles were captured on traps baited with Df phloem than those baited with Df bark, sapwood or heartwood. These results indicate which tissues might be extracted to determine the identity of primary attractants. It was concluded that WH i s preferred as a host tree over Df by G_. sulcatus during the f i r s t year, but both hosts had similar number of attacks in the second year. Such preference for WH poses a problem for logging managers in coastal British Columbia since WH is 40% of the total volume processed. If G. sulcatus populations are allowed to build up in stumps, they could degrade logs in felled and bucked timber in adjacent areas. WH should be removed as soon as possible after f e l l i n g (hot logging) and the period in dryland sorts should be minimized as logs remain vulnerable to attack 17 months following f e l l i n g . i i i DEDICATION This thesis is dedicated to my wife, Teresinha Vinha Zanuncio, who encouraged me to come to Canada and made a l l the efforts and sacrifices to help me while here and also to my daughter Andressa Vinha Zanuncio. ACKNOWLEDGEMENTS I express my deepest gratitude to the following persons and organizations without whose contribution this thesis would not be possible. Dr. J.A. McLean, my supervisor, for the invaluable efforts to introduce me to this topic, firm guidance and help; my committee members, Drs. J.H. Borden, R.H. E l l i o t t , B.J. van der Kamp and G.F. Weetman for their helpful suggestions and advice; Dr. A. Kozak for helping and recommending the s t a t i s t i c a l analysis used. J. Bebyck, J. Holman and M. J u l l for their help in the f i e l d and laboratory work; and Mr. P. Saunders of the UBC Research Forest for f e l l i n g and cutting trees used in this study. Also Mr. A. Magnuson for welcoming research on his selectively-logged forest at Point Roberts. The Canadian International Development Agency (CIDA) and the Federal University of Vicosa (UFV) for financial support. I thank Dr. V.J. Nordin and Mr. A.F. Shirran for their help through CIDA in Canada and also Dr. Arno Brune, Coordinator of the CIDA/UFV agreement in Brazil. V TABLE OF CONTENTS Page ABSTRACT i DEDICATION i i i ACKNOWLEDGEMENTS iv TABLE OF CONTENTS v LIST OF TABLES v i i i LIST OF FIGURES x LIST OF APPENDICES x i i i 1.0 INTRODUCTION 1 2.0 MATERIAL AND METHODS 9 2.1 Field Study Locations and Experimental Designs 9 2.1.1 UBC Research Forest 9 2.1.2 Point Roberts Host Tissue Attraction Studies .. 11 2.2 Host Colonization 12 2.2.1 Verification of Attacks 12 2.2.2 Weekly Pinning 12 2.2.3 Distribution Around Logs and Pinning Efficiency 12 2.3 Brood Production by Gnathotrichus sulcatus on Stumps, 13 Logs and Downtrees • 13 2.4 Gallery Dissection 16 2.5 Laboratory Rearing 16 2.6 Primary Attractants Tests 18 2.6.1 Test with Western Hemlock Stimuli 18 2.6.2 Test with Douglas-fir Stimuli 21 v i Page 2.7 Data Analysis 22 2.8 Total Brood Emergence Pattern Over Time 22 3.0 RESULTS 24 3.1 Host Colonization by Gnathotrichus sulcatus 24 3.1.1 Attacks on Each Host Category 24 3.1.2 Comparison of Attack Densities 24 3.1.3 Seasonal Attack Rates 30 3.1.4 Attack in Relation to Moisture Content in Trunks and Stumps 30 3.1.5 Distribution Around Trunks 34 3.1.6 Evaluation of Pinning Efficiency 34 3.2 Host Colonization by Associated Scolytid Species .... 36 3.2.1 Dendroctonus pseudotsugae, Hopk., the Douglas-f i r Beetle 36 3.2.2 Scolytus unispinosus LeConte, the Douglas-fir Engraver 36 3.2.3 Scolytus tsugae (Swaine), the Hemlock Engraver 38 3.2.4 Pseudohylesinus tsugae Swaine 38 3.2.5 Trypodendron lineatum (Olivier), the Striped Ambrosia Beetle 40 3.3 Establishment Success of Gnathotrichus sulcatus 40 3.4 Frass Production 43 3.5 Brood Production 47 3.5.1 Stumps in the Field 47 3.5.2 Trunks in the Field 48 v i i Page 3.5.3 Logs i n Laboratory 52 3.5.4 Sex of Emerging Beetles 56 3.6 Gallery Dissections 56 3.6.1 Stumps 56 3.6.2 Trunks 61 3.6.3 Relationships between Brood, Frass and Length of Galleries 62 3.6.4 Aspects of Gallery Construction by Gnathotrichus sulcatus at Different Periods of Time after Felling 62 3.7 Primary Attractants Tests 68 3.7.1 Western Hemlock Stimuli 68 3.7.2 Douglas-fir Stimuli 72 3.8 Total Brood Emergence Pattern 72 4.0 DISCUSSION 76 4.1 Host Selection and Colonization by Gnathotrichus sulcatus 76 4.2 Host Selection and Colonization by Associated Scolytid Beetles 81 4.3 Establishment and Gallery Parameters 83 4.4 Aspects of Gallery Construction of Gnathotrichus sulcatus in Trunks for Different Periods after Felling 87 4.5 Relationship between Brood Emergence Pattern and Attack Data 89 4.6 Implications of the Results in B.C. Forest Industry 94 5.0 SUMMARY 96 LITERATURE CITED 98 APPENDICES 107 v i i i LIST OF TABLES Table I A review of observations on primary attraction in Gnathotrichus sulcatus. Page Table II Length and diameter at breast height (d.b.h.) for western hemlock and Douglas-fir logs and downtrees felled in the UBC Research Forest, May 8, 1978. 10 Table III Total number of Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps, logs (L) and downtrees (DT). UBC Research Forest, Maple Ridge, B.C. 1978/79. 25 Table IV Summary of attack densities of Gnathotrichus  sulcatus on western hemlock and Douglas-fir stumps, logs and downtrees. UBC Research Forest, Maple Ridge, B.C. 1978-79. 27 Table V Bi-weekly mean attack/m^ Q f Gnathotrichus sulcatus on western hemlock and Douglas-fir logs and downtrees. UBC Research Forest, Maple Ridge, B.C. 1978/79. 28-29 Table VI Distribution of Gnathotrichus sulcatus attacks by quadrants on 1 m sections of western hemlock and Douglas-fir logs. UBC Research Forest, Maple Ridge, B.C. May-November, 1978 and March-October, 1979. N = 8 for each species. 35 Table VII Number and success of Gnathotrichus sulcatus attacks monitored with petri dishes on western hemlock stumps and trunks and Douglas-fir stumps and trunks in 1978 and 197 9. UBC Research Forest, Maple Ridge, B.C. 42 Table VIII Mean total frass production from 147 galleries of Gnathotrichus sulcatus monitored on western hemlock and Douglas-fir stumps and trunks in the UBC Research Forest, Maple Ridge, B.C. 1978-79. 44 Table IX Brood production from 147 galleries of Gnathotrichus sulcatus monitored on western hemlock and Douglas-fir stumps and trunks in the UBC Research Forest, Maple Ridge, B.C. 1978-79. 47 Total numbers of brood produced and sex ratio from Gnathotrichus sulcatus galleries established in western hemlock and Douglas-fir under laboratory conditions, UBC 1979. Sex ratio of f i r s t and total Gnathotrichus  sulcatus emerging from western hemlock and Douglas-fir trunks and stumps felled May 1978 in the UBC Research Forest, Maple Ridge, B.C. Mean egg niches (EN), pupal niches (PN), length of gallery (LG) and depth of penetration (DP) of Gnathotrichus sulcatus galleries on western hemlock and Douglas-fir stumps and trunks. UBC Research Forest, Maple Ridge, B.C. Relationships between brood, frass and length of galleries of Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps and trunks in the UBC Research Forest, Maple Ridge, B.C. Egg niches, pupal niches, length of galleries and depth of penetration of Gnathotrichus sulcatus on western hemlock and Douglas-fir trunks over 13 months, UBC Research Forest, Maple Ridge, B.C. 1978-79. Number of Gnathotrichus sulcatus galleries with and without pupal niches, and total number of pupal niches on western hemlock and Douglas-fir logs over 13 months. UBC Research Forest. 1978/79. Numbers of Gnathotrichus sulcatus caught on traps baited with stimuli prepared from western hemlock bark, sapwood and heartwood; and from Douglas-fir bark, phloem, sapwood and heartwood. Experiment set out in a selectively logged forest in Point Roberts, WA on June-July, 1978. Brood production profile from the estimated 16,049 Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps and trunks from June, 1978 to June, 1980. UBC Research Forest, Maple Ridge, B.C. X LIST OF FIGURES Page 14 15 17 20 31 Figures 1, 2 Pinned attacks of Gnathotrichus sulcatus and petri dishes to capture brood in successful galleries. 1, WH stump; 2, WH trunk. Figure 3 Petri dishes on western hemlock showing dark colour of frass five weeks after attack. Figure 4 Log of Douglas-fir in laboratory showing the five petri dishes, the water container on the top and the water reservoir at the bottom of the log. Figure 5 Modified survey trap used for testing host stimuli i n Point Roberts in 1978. Figures 6, 7 Bi-weekly totals of pinned attacks of Gnathotrichus sulcatus on western hemlock and Douglas-fir. UBC Research Forest. May-November/1978 and March-October/1979. 6, Stumps; 7, Logs and downtrees. Figures 8, 9 Climate data recorded at the gate of the UBC Research Forest from May 18, 1978 to December 31, 1979. Maple Ridge, B.C. 8, Weekly r a i n f a l l ; 9, Minimum and maximum weekly temperature. Figures 10, 11 Mean moisture content of western hemlock and Douglas-fir felled May 8, 1978 in the UBC Research Forest for the periods May 1978 -November 1979. Initiation of sustained attacks by Gnathotrichus sulcatus indicated by solid arrows and f i r s t attack indicated by wavy arrows. Numbers refer to host pairs. Vertical bars represent standard errors of reported means, 10, Stumps; 11, Logs and downtrees. Figures 12, 13 Seasonal attack pattern of two bark beetles on Douglas-fir at the UBC Research Forest in 1978. Maple Ridge, B.C. 12, Dendroctonus pseudotsugae; 13, Scolytus unispinosus 37 Figures 14, 15 Seasonal attack pattern of two bark beetles on western hemlock at the UBC Research Forest in 1978. Maple Ridge, B.C. 14, Scolytus tsugae; 15, Pseudohylesinus tsugae. 39 32 33 Figure 16 Figures 17, 18 Figures 19, 20 Figures 21, 22 Figures 23, 24 Figures 25-27 Figures 28-31 Total bi-weekly numbers of Trypodendron  lineatum attacks pinned on western hemlock and Douglas-fir in 1979 at the UBC Research Forest, Maple Ridge, B.C. Mean weekly frass production of galleries initiated at various times during 1978 and 1979 by Gnathotrichus sulcatus on western hemlock, at the UBC Research Forest. 17, Logs and downtrees, A - 5 attacks from June 2-June 30, B - 27 attacks from July 14-September 29, C - 5 attacks from May 1-June 5, D - 4 galleries on May 15; 18, Stumps, A - 26 attacks from June 2-July 7, B - 6 attacks from May 1-May 22, C - 6 galleries from May 22-June 5. Mean weekly frass production of galleries initiated at various times during 1978 and 1979 by Gnathotrichus sulcatus on Douglas-fir, at the UBC Research Forest. 19, Df logs and downtrees, A - 8 attacks from Sept. 1-Sept. 22, B - 6 attacks from May 1-May 22; 20, Stumps, A - 5 attacks from July 28-Sept. 1; B - 11 attacks from May 15-June 5, 1979. Total brood production from Gnathotrichus  sulcatus galleries monitored in the UBC Research Forest from June 1978 to June 1980, Maple Ridge, B.C. 21, Site 1, (N = 67); 22, Site 2 (N = 80). Mean weekly brood production from Gnathotrichus  sulcatus galleries established on western hemlock in June - early July (early summer). UBC Research Forest, Maple Ridge, B.C. BP50 and BP90 are the f i f t y and ninety percentiles for cumulative brood production. 23, WH trunks; 24, WH stumps. Mean weekly brood production from Gnathotrichus  sulcatus galleries established on WH and Df from late July to October, 1978, UBC Research Forest, Maple Ridge, B.C. 25, WH stumps; 26, Df trunks; 27, Df stumps. Mean weekly brood production from Gnathotrichus  sulcatus galleries established on WH and on Df in May-June, 1979. UBC Research Forest, Maple Ridge B.C. 28. WH logs and downtrees. 19, WH stumps; 30, Df logs and downtrees; 31, Df stumps. Figure 32 Cumulative mean Gnathotrichus sulcatus brood production in western hemlock and Douglas-fir logs in laboratory conditions, Faculty of Forestry, UBC. Numbers in parentheses indicate number of successful galleries in each host log. Figure 33 Schematic plan of Gnathotrichus sulcatus gallery in western hemlock two years after attack. Note the presence of egg niches (EN), pupal niches (PN), dead beetles (D), and live beetles (A). Figure 34 Mean egg niche, pupal niche and brood produc-tion of Gnathotrichus sulcatus galleries on WH stumps and trunks and on Df stumps and trunks. UBC Research Forest, Maple Ridge, B.C. Figures 35, 36 Mean number of egg and pupal niches of Gnathotrichus sulcatus on western hemlock and Douglas-fir trunks from June 1978 to October 1979. UBC Research Forest. 35, Egg niches; 36, Pupal niches. Figures 37, 38 Gnathotrichus sulcatus gallery parameters on western hemlock and Douglas-fir trunks from June 1978 to October 1979. UBC Research Forest. 37, Length of galleries; 38, Depth of penetration. Figure 39 Estimates of total monthly attacks and brood production during 1978 through 1980 for western hemlock and Douglas-fir host material felled in May 8, 1978. UBC Research Forest, Maple Ridge, B.C. x i i i LIST OF APPENDICES Appendix Page Results of the analysis of variance of Gnathotrichus  sulcatus attacks/m^ on western hemlock and Douglas-f i r stump in the UBC Research Forest, 1978-79. Results of the analysis of variance for Gnathotrichus  sulcatus attacks/m^ on western hemlock and Douglas-fir trunks in the UBC Research Forest, 1978-79. Results of the analysis of variance for percent of Gnathotrichus sulcatus attacks per quadrant on 1 m long western hemlock and Douglas-fir. UBC Research Forest, Maple Ridge, B.C. May-November, 1978 and March-October, 1979. N = 8 for each species. Results of the analysis of variance of bark beetles total attacks on western hemlock and Douglas-fir trunks in the UBC Research Forest, 1978-79. 107 108 109 110 5 Results of the analysis of variance of Trypodendron  lineatum attacks/m2 on western hemlock and Douglas-f i r trunks in the UBC Research Forest, 1978-79. I l l 6 Results of the analysis of variance of brood (BR) and frass (FR) of Gnathotrichus sulcatus galleries in western hemlock and Douglas-fir initiated during 1978-79. UBC Research Forest, Maple Ridge, B.C. 112 7 Results of the analysis of variance for Gnathotrichus  sulcatus productivity/gallery and mean gallery parameters in western hemlock and Douglas-fir logs in laboratory conditions, Faculty of Forestry, UBC 197 9". 113 8 Results of the analysis of variance of mean numbers of egg niches (EN) and pupal niches (PN) of Gnathotrichus sulcatus galleries on western hemlock and Douglas-fir initiated during 1978-79. UBC Research Forest, Maple Ridge, B.C. 114 9 Results of the analysis of variance of mean length (LG) and depth of penetration (DP) of Gnathotrichus  sulcatus galleries on western hemlock and Douglas-f i r initiated during 1978-79. UBC Research Forest, Maple Ridge, B.C. 115 Results of the analysis of variance of mean numbers of egg niches (EN), pupal niches (PN), depth of penetration (DP) and length (LG) of Gnathotrichus  sulcatus galleries in western hemlock trunks over 13 months. UBC Research Forest, Maple Ridge, B.C. 1978-79. Results of the analysis of variance of egg niches (EN), pupal niches (PN), depth of penetration (DP) and length (LG) of Gnathotrichus sulcatus galleries in Douglas-fir trunks over 13 months. UBC Research Forest, B.C. 1978-79. Results of the analysis of variance for Gnathotrichus sulcatus catches on bark, phloem, sapwood and heartwood stimuli prepared from western hemlock and Douglas-fir. Set out in Point Roberts, Washington on June and July, 1978. 1 1.0 INTRODUCTION Ambrosia beetles typically colonize recently dead or felled trees i n the coniferous forests of the Pacific Northwest. They are the f i r s t agent of wood decomposition and they speed recycling of nutrients in the forest. Prior to man's logging a c t i v i t i e s , the events which maintained supplies of suitable host material included wildfires, wind blow down, snow press, flooding and ice breakage. These beetles comprise a l l the Platypodidae and many Scolytidae, well over 1000 species in a l l . They vary in length from l i t t l e more than 1 mm to over 10 mm in the longest Platypodidae species (Anonymous 1972). Five species of ambrosia beetles commonly occur in the Pacific Northwest. Among them, Gnathotrichus sulcatus (LeConte), together with i t s associated species, Trypodendron lineatum (Olivier), G_. retusus LeConte, Platypus wilsoni Swaine, and Xyleborus saxeseni (Ratz.), cause considerable economic loss by degrade of wood products (McBride and Kinghorn 1960). Degrade of wood in B.C. was estimated at around 11% of total volume in 1948 (McBride 1950), and the value of better classes of timber has been reported as being reduced by 50% (Graham and Boyes 1950). Nijholt (1978) stated that in B.C., the damage due to ambrosia beetles in 1975-76 was estimated by the Western Forest Products Laboratory, Vancouver, B.C.*, to be around $7 million. 1 Now Forintek Canada Corp., Western Laboratory. 2 Additional industry losses are associated with increased slabbing and edging waste and reduced mil l output per unit of log manufactured. Thirdly, there i s the effect of increased manufacturing time on the headrig when the sawyer attempts to establish a clean face on a log .before passing i t on the edger (Graham and Boyes 1950). Finally, interceptions of wood containing ambrosia beetles such as G_. sulcatus have caused quarantine problems in timber importing countries such as New Zealand (Milligan 1970; Bain 1974). In extreme cases, wood shipments to Australia were ordered destroyed in 1928 and 1930 due to ambrosia beetles infestations (Mathers 1935). Such precautions are advisable since McLean and Borden (1975a) have demonstrated that G_. sulcatus can survive and reproduce in freshly sawn lumber, thus showing the very real possibility of this ambrosia beetle being introduced into foreign markets. An example of ambrosia beetle impact in a new environment is demonstrated by X. saxeseni which was accidentally introduced i n New Zealand (Milligan 1969) and now infests some 31 species of native trees (Hosking 1972). For efficient management of a pest such as G_. sulcatus, the understanding of i t s biology i s of foremost importance. This species does not appear in large numbers in a single major spring flight period as does the associated holoartic species, T_. lineatum. Most ambrosia beetle research in B.C. has previously concentrated on T_. lineatum (Nijholt 1979). Observations on G. sulcatus were sometimes included in publications as incidental notes. Daterman et a l . (1965) stated that since much of the l i f e history of this insect is unknown, i t was 3 impossible to define the seasonal flight in terms of origin of beetles in flight during the season. Several other authors, indicated a bi-modal flight pattern, a peak in the spring or early summer and a second in September (Prebble and Graham 1957; Chapman and Kinghorn 1958; McLean and Borden 1979), but no explanation for such bi-modality - has been given. Rudinsky and Schneider (1969) suggested that were two generations per year when they found CJ. sulcatus starting galleries in October. G_. sulcatus was f i r s t described by LeConte in 1868 as Crypturgus  sulcatus along with two other species, C^. retusus and £. asperubus. In the same year Eichhoff created the genus Gnathotrichus for (J. cortyloides, which he believed could be Tomiscus materiarius. Three other new species of Gnathotrichus from Chile were also described at that time (Blackman 1931). In 1895, Blandford redescribed the genus Gnathotrichus and included two new species from Central America. Later in 1918, Swaine put not only T_. materiarius but also (3. retusus LeConte and C_. sulcatus LeConte in this genus (Blackman 1931). Doane j2t a l . (1929) presented the f i r s t data on G. sulcatus biology on Douglas-fir (Df), Pseudotsuga menziesii (Mirb.) Franco, when they stated in California that the beetles attacked the top and then the trunk of the tree, and later could also be found in the stumps where they continued their borings for several generations. They further conjectured that copulation took place in the main and secondary galleries and that eggs hatched in about 7 or 8 days. Doane et a l . (1936) stated that G. sulcatus in California reproduced and continued to extend galleries as long as the wood of their host was 4 sufficiently moist for the growth of the associated fungi. They also usually found two or more broods each year, but these were not regular, since larvae, pupae, and adults were present within the galleries at almost any season. Beetles were in flight from the middle of April to the f i r s t week of September and they apparently preferred fresh timber (Doane et a l . 1936). Age of host material varied greatly, with several logs being attacked within a week of the time of fe l l i n g and one log 15 months after i t was cut (Mathers 1935). Most of the more recent studies on G_. sulcatus have concentrated on primary and secondary attraction. G_. sulcatus started to emerge and fly in the spring when temperatures exceeded 16°C (Daterman et a l . 1965). Rudinsky and Schneider (1969) demonstrated that temperature and light intensity regulated maximum flight activity. Once in f l i g h t , G_. sulcatus was guided to new habitats by odours produced by suscep-tible host tissues (Graham and Werner 1956; Chapman 1963). Person (1931) hypothesized that fermentation products from dying trees were responsible for the attraction of scolytid beetles which subsequently attacked and colonized them. While the hypothesis has been substan-t i a l l y disproved for the bark beetle, Dendroctonus brevicomis LeConte for which i t was originally erected (Moeck et a l . 1980), i t appears to be valid for ambrosia beetles as proposed by Doane et a l . (1936). Ethanol has been identified as one of the primary attractants (Cade et a l . 1970); i t i s hypothesized as being formed by anaerobic respiration in host tissues (Graham 1968). Low numbers of G_. sulcatus can be captured at traps baited with ethanol (Cade et a l . 1970; Moeck 1971; 5 McLean 1976). Significantly more beetles were captured on traps baited with ethanolic solutions of a-pinene than on ethanol-baited traps (Rudinsky 1966). Observations related to Gnathotrichus primary attrac-tion are reviewed in Table 1. The presence or absence of the needles on downtrees (Ford-Robertson 1971) might influence primary attraction. It has been suggested that downtrees were attacked less densely than logs by G_. sulcatus (Johnson 1964). However, an experiment to test the hypothesis (Johnson and Zing 1969) failed to demonstrate any s i g n i f i -cant differences between attack rates on these two categories of host. Isolation procedures for identifying the primary attractants for G_. sulcatus could be greatly enhanced by demonstrating which host tissue i s the most attractive. McLean (1976) showed greatest responses by walking beetles in a laboratory olfactometer to air passing over sapwood sawdust prepared from western hemlock (WH), Tsuga heterophylla (Raf.) Sarg., than to air alone. Ambrosia beetles feed on ambrosia fungi carried by the adults when they emerge and f l y to a host. As the ambrosia fungi are trans-ported by female beetles to new tunnels during excavation and before oviposition, adults do not consume ambrosia fungus at this time but feed on wood as they do during the maturation feeding (Batra 1963). Farris (1963) found that male G. sulcatus carry their symbiotic fungi in special coxal cavities of the prothoracic legs; seasonal changes in these mycetangial glands were described by Schneider and Rudinsky (1969). These fungi were described by Funk (1970) as Ambrosiella  sulcati Funk, Raphaella sulcati Funk and Graphium sp. TABLE I. A review of observations on primary a t t r a c t i o n In Gnathotrichus sulcatus Observation Reference A. General Concepts: - Primary attractants produced by fermentation of host tissue (western pine beetle) - Sap must be in fermentatlng condition for attack (ambrosia beetles) - Primary attractants produced by anaerobic respiration of host tissues deprived of translocatlve oxygen B. Response to Host Tissue (odors) and Primary Attractants ~ Host Tissues . Ethanol was i d e n t i f i e d as a primary attractant for G. sulcatus. Catches to treatments were: soaked hemlock (31), 80Z ethanol (22), 50% ethanol and non-soaked hemlock (10), water (0). Ethanol Identified as the major v o l a t i l e In soaked hemlock. In laboratory bloassays either western hemlock sawdust or ethanol alone had a higher response for males and females of G. sulcatus than a i r control. When tested together the response was higher than to either attractant, alone (additive response) - Ethanol alone 282 males and 58 females were captured in 10Z ethanol-baited traps as opposed to 7 males and 9 females i n water only control traps Higher response by G. sulcatus to ethanol than control In walking bloassays. Traps baited with ethanol with a release rate of 10.0 and 35.1 g/24 hr captured more beetles than unbalted control . G_. sulcatus response to ethanol i n a sawmill study was not s i g n i f i c a n t l y greater than to unbalted control (test included sulcatol-baited traps) . Response of G. sulcatus to ethanol was not s i g n i f i c a n t l y different from the unbalted control (test Included a sulcatol treatment) - Ethanol + host compounds • G. sulcatus had the higher response to, i n descending order, ethanolic solutions of a-pinene, camphene and B'pinene. No response to ethanol-baited traps Ethanol-balted logs had greater numbers of attacks than control logs In f i e l d experiment • Ethanol + a-plnene captured s i g n i f i c a n t l y more beetles than control C. - Primary Attractants plus Pheroaones - S i g n i f i c a n t l y more attacks occurred in western hemlock logs and Douglas-fir logs and stumps baited with sulcatol + ethanol than to host material sulcatol-baIted alone - S i g n i f i c a n t l y more males responded to traps baited with (+) sulcatol + ethanol + ct-pinene than to traps baited with (+) sulcatol alone Person 1931 Doane et a l . 1936 Graham 1968 Cade et a l . 1970 McLean 1976 Hoeck 1971 McLean 1976 McLean & Borden 1975a Borden et a l . 1980a Rudinsky 1966 McLean & Borden 1977 Borden et a l . 1980b McLean & Borden 197 7a Borden et a l . 1980a 7 After successful establishment of G. sulcatus in a suitable host, secondary attraction (Borden 1974) i s set up. This was demonstrated in the laboratory (Borden and Stokkink 1973). An aggregation pheromone present in the boring dust of male beetles (Borden and Stokkink 1973) was isolated, identified as 6-methyl-5-hepten-2-ol and given the t r i v i a l name sulcatol (Byrne et a l . 1974). In fi e l d t r i a l s , traps baited with a racemic mixture of synthetic sulcatol attracted large numbers of both sexes of G_. sulcatus despite competition of natural host attractants (Byrne et j i l . 1974). Primary attractants enhance the catches at traps baited with sulcatol (Borden et a l . 1980a) and they have a significant role in secondary attraction. The related species G. retusus u t i l i z e s 5-(+)-sulcatol as it s aggregation pheromone (Borden et a l . 1980c). The objectives of this work were to: 1. Determine the natural colonization patterns of G. sulcatus on WH and Df stumps, logs (L) and downtrees (DT) and record attacks by associated scolytids at the UBC Research Forest over two successive years. 2. Determine when peak attack occurs especially as related to vulnerability of May-felled logs and whether there is any differential attack rate between logs and downtrees. 3. Determine frass and brood production from these attacks by placing emergence chambers over successful galleries on each host type and to establish when brood emerged and the relationship between attacking and emergent beetles. 8 4. Correlate frass and brood production with length of gallery to derive a developmental success profile. 5. Determine the success of galleries for each month of attack. 6. Determine the host tree tissue which is most attractive to flying beetles. 9 2.0 MATERIALS AND METHODS 2.1 FIELD STUDY LOCATIONS AND EXPERIMENTAL DESIGNS 2.1.1 U.B.C. Research Forest The research on natural colonization, brood production and effect of log age on gallery success was done in the U.B.C. Research Forest, Maple Ridge, B.C., 60 km east of Vancouver. This forest i s located in the coastal western hemlock biogeoclimatic zone (Klinka 1976). The two study sites were located at altitudes of 110 m and 122 m in second growth f i r e succession stands in a Rubus - Polystichum - western red cedar ecosystem. Four pairs of trees were felled on each site on May 8, 1978. Each pair consisted of one Douglas-fir (Df) and one western hemlock (WH). At each site branches were l e f t intact on two pairs of felled trees (downtrees), while branches were removed on the other two pairs (logs). Length, d.b.h. and other characteristics of these logs and downtrees are listed in Table II. The 2 x 2 x 2 factorial design, which included two sites, two hosts and two conditions with two replications per site/host/condition/ situation, was used to study natural colonization patterns of Df and WH by ambrosia and bark beetles. The experimental design enabled compari-son to be made between attack patterns on WH and Df stumps and between logs and downtrees of the two host species at the two sites. Brood and frass production was recorded for five galleries in each stump, log and downtree. The effect of host age on the success of 10 TABLE II. Length and diameter at breast height (d.b.h.) for western hemlock and Douglas-fir logs and downtrees, Felled in the UBC Research Forest, May 8, 1978 Host species Site Tree # Condition Length (m) d.b.h. (cm) Age (yrs) Western Hemlock 1 2 3 A 5 6 7 8 DT 14.0 23.7 59 L 22.0 34.6 70 DT 13.0 32.2 67 L 12.0 26.9 73 DT 23.0 42.8 69 L 25.0 38.1 71 DT 26.0 27.8 71 L 14.0 29.1 69 Douglas-fir 1 DT 17.0 26.0 72 2 L 17.0 30.9 74 3 DT 27.0 31.9 78 4 L 27.0 44.5 73 5 DT 27.0 38.5 85 6 L 23.0 17.2 71 7 DT 24.0 26.8 70 8 L 25.0 27.9 73 *DT = downtree (with branches), L = log (no branches). 11 G_. sulcatus attack was evaluated by dissecting two galleries initiated in each month of this study for each host category. Wood samples from each stump and from the third meter of each log and downtree were taken weekly using an increment borer. The percentage moisture on a dry weight basis of each sample determined on return to the laboratory. Samples for moisture analysis were always taken from the same side and area to avoid any variability that might be associated with sunny and shady sides of the host material. Weekly ra i n f a l l and temperature records were obtained from the meteorological station maintained by Research Forest personnel. 2.1.2 Point Roberts Host Tissue Attraction Studies The tests of WH and Df stimuli were set out in a selectively logged coastal Douglas-fir wetter subzone forest in Point Roberts, Washington, U.S.A., some 30 km south of Vancouver. This kind of forest operation maintains a constant supply of forest host material for the beetle population on a year-by-year basis. This experiment was installed in five blocks, each block being set out along a logging skid road. A Latin square randomization pattern was used to ensure that each treatment was assigned to a different position in each block, since traps in the ends of a row tended to catch more beetles (McLean and Borden 1977b). 12 2.2 HOST COLONIZATION 2.2.1 Verification of Attacks The identities of attacking scolytid beetles were verified by excavating new attacks until symptoms were reliably recognized. Such excavations were done weekly through the end of the experiment to avoid misidentification of species. Ambrosia beetles monitored included £. sulcatus and T. lineatum. Bark beetles monitored included D. pseudotsugae Hopk., Pseudohylesinus tsugae Swaine, Scolytus tsugae (Swaine) and S^. unispinosus LeConte. Samples of each species were sent to the Biosystematics Research Institute, Ottawa to verify identifica-tions. 2.2.2 Weekly Pinning Host material was checked weekly and fresh attacks pinned from May 19 to November 2, 1978 and from March 25 to October 30, 1979. Each week, pins were color coded so that a temporal sequence of attack could be analysed in any area. 2.2.3 Distribution Around Logs and Downtrees and Pinning  Efficiency One meter sections were removed from each log and downtree around the third meter from the butt end at the conclusion of the study to determine distribution of attacks on top, sides and bottom (quadrant) of the logs and downtrees, and the opportunity was also taken to 13 evaluate pinning efficiency for G_. sulcatus. On the stumps pinning efficiency was determined on the disks taken to assess brood production. 2.3 BROOD PRODUCTION BY Gnathotrichus sulcatus ON STUMPS, LOGS AND DOWNTREES Frass and brood production in successful G_. sulcatus galleries were monitored by covering gallery entrances with plastic petri dishes (Hosking 1972). The base of the 5.5 cm diameter petri dish had a central hole which was placed over the gallery entrance. To i n s t a l l the dish, bark was f i r s t smoothed with a chisel and a small quantity of plasticine was used to ensure a seal between the dish and the trunk or stump. Dishes were nailed to the bark and covered after 1 week. The delay was to allow for pair formation. The cover was held in place with a small rubber band, which was stretched over the cap between two nails close the edge of the petri dish (Fig. 1). Attacks were covered in each host category until 5 successful galleries were obtained in a shortest period of time after the f i r s t attack on the stump, log or downtree (Fig. 2). Since attack rate was far from uniform and did not happen at the same time for every stump, log and downtree, the dish installation f i r s t started in June 1978 and extended to July 1979 when most host categories were fi n a l l y attacked. One Df stump was never attacked. A l l frass and brood were collected weekly until production ceased (Fig. 3). They were collected in a small v i a l in the fie l d and brought to UBC, where the beetles were sexed and the frass dried at 14 Figures 1, 2: Pinned attacks of Gnathotrichus sulcatus and p e t r i dishes to capture brood i n successful g a l l e r i e s . 1, WH stump; 2, WH trunk. 15 Figure 3: Petri dish on western hemlock showing dark colour of frass five weeks after attack. 16 100°C for 24 hours, allowed to cool down for one hour in a dry desic-cator and then weighed. The beetles that came out from galleries during the f i r s t week of emergence and total production from a successful gallery were studied to determine i f there was any predominance of either sex as the f i r s t to emerge. 2.4 GALLERY DISSECTION During the spring and summer of 1980, 10 cm thick disks were sawn from each stump, log and downtree so that the complete galleries developed under a l l petri-dish covered attacks could be brought into the laboratory. Each gallery was dissected and a drawing made to show locations of egg and pupal niches, depth of penetration and length of galleries, as well as any dead insects. In addition, two attacks from the months of June through October, 1978 and from March to October 1979, in each WH and Df log and downtree host category were dissected. 2.5 LABORATORY REARING Five pieces of log were brought to UBC, stored outside for four months and then kept in a chamber at 23°C. A constant 14 hr light/10 hr dark regime was maintained in the chamber. Logs were placed on end in a shallow tray of water. A perforated plastic 500 ml container of water on the upper cut surface allowed water to trickle down over the log thereby maintaining a high moisture regime (Fig. 4). The water was replenished whenever necessary. 17 Figure 4: Log of Douglas-fir i n laboratory showing the fi v e p e t r i dishes, the water container on the top and the water reservoir at the bottom of the l o g . 18 Several 3 mm diameter and 2 cm deep holes were dr i l l e d in each ~ log, a petri dish was installed over each hole and a male was put in each gallery. Four to five days later a female beetle was added. A l l parent beetles used in this experiment were brood collected in the f i e l d work. The resulting brood were collected daily in the laboratory. After development was completed, a l l successful and unsuccessful galleries were dissected and a drawing made of each one to show length, depth of penetration and positions of pupal and egg niches. The sex ratio of first-emerging beetles and of the total brood were determined. The number of degree-days required for development was calculated from the day when a female joined a male in a successful gallery until the appearance of the f i r s t brood. This method is recommended when temperature is relatively stable (Mizzel and Nebecker 1978). For the f i e l d work which was carried out under normal fluctuating diurnal temperatures, the number of degree-days was calculated from the attack date to the average date when half the brood had been produced in each of 37 galleries (Mizzel and Nebecker 1978). 2.6 PRIMARY ATTRACTANTS TESTS 2.6.1 Test with Western Hemlock Stimuli On June 2, 1978 1 m lengths of WH were cut from each of six different logs felled between May 16 and May 19 in the UBC Research Forest near Maple Ridge, B.C. Logs from six trees were taken because of the high variation in susceptibility between trees within the same 19 species (Prebble and Graham 1957; Chapman 1962). The six pieces of log were brought to UBC, and on June 4, 1978, the bark was stripped off, and sapwood and heartwood sawdust was prepared using a chainsaw. The bark was reduced to small pieces, approximately 1 cm^ , with a machete. Bark pieces and sapwood and heartwood sawdust from the 6 logs were each combined and stored in plastic bags at 12.8°C and 74% R.W. These host materials were packed in cylinders 10 cm in diameter and 22 cm long made of 6 mm wiremesh (Fig. 5). A 1.5 m high pole was put through each cylinder which was fixed at 10 cm from the top of the pole. Each cylinder of stimulus was surrounded by a wiremesh cylinder trap (Byrne et a l . 1974) coated with Stikem Special 1 (Fig. 5). Empty cylinders in the control traps were wrapped in brown paper to provide the same silhouettes as were present i n the other traps. Field experiments were set out along logging spur roads in a selectively logged forest in Point Roberts, WA on June 7, 1978. The treatments were: bark, sapwood, heartwood, sulcatol and blank. Each block was set out along a logging skid road with approximately 30 m between traps. The sulcatol treatment was included for 1 week to provide a check on the presence of beetles in the test area, and was then removed so as not to overwhelm any host odour stimulus. Michel PeltonLtd., Emeryville, CA. 20 .20*5 cm PLYWOOD SUPPORT DISCS WIREMESH CYLINDER TO ENCLOSE STIMULI WIREMESH CYLINDER COVERED WITH STIKEM SPECIAL® SUPPORT POLE 1-5m HEIGHT Figure 5: Modified survey trap used for testing host stimuli in Point Roberts in 1978. I 21 The experiment ran for five weeks, without changing the stimuli. (5. sulcatus were collected from the traps each week, and brought to the laboratory and their sex were determined. Two further experiments to test attraction of bark, phloem, sapwood and heartwood of WH were done in July 1979. 2.6.2 Test with Douglas-fir Stimuli On July 7, 1978, two Df were felled at the UBC Research Forest and three pieces, each approximately 1 m long, were taken from the butt end of each log. Sapwood and heartwood sawdust and bark stimuli were prepared and stored as for those from WH. The only difference was that an additional stimulus, phloem, was prepared since i t was easily separated from the bark. On July 12, the wiremesh cylinders were put on a trap support pole and f i l l e d with host stimuli. Following this, they were wrapped in plastic bags and kept in the cold chamber until June 16, 1978, when the experiment was set up at Point Roberts, WA. Experimental design, beetle collection and s t a t i s t i c a l analyses were the same as for WH, except that the stimuli were bark, phloem, sapwood, and heartwood. Sulcatol-baited traps were not set out. The experiment ran for three weeks, but fin a l analyses were done for catches during the f i r s t week only as too few beetles were captured in weeks 2 and 3. 22 2.7 DATA ANALYSIS Prior to analysis of variance, the data were transformed X' = log^Q (X + 1), except in the case of data for the percent of attacks by quadrant, where the raw data were used since the variances were 2 homogeneous. Data were analysed using the package ANOVAR and when 3 there were missing data, the GENLIN package was used. In a l l cases the Newman Keul's Test (Winer 1971) was used to test ranked means within main effects. For the regression and correlation analyses the 4 MIDAS package was used. 2.8 TOTAL BROOD EMERGENCE PATTERN Estimates of total attacks and total numbers of brood on the stump/trunk combinations at the UBC Research Forest were calculated from data derived in this study i n order to show the contribution of each attack period to the total brood production. This was done as follows: The numbers of pinned attacks (NA) (Table III) were totalled by month for each host category and then converted to an estimated of total attacks (TA) using the pinning efficiency (PE) calculated in Section 3.1.6. TA - NA/PE (1) The TA were converted to a proportion of attacks with pupal niches (PPN). This was done by each month for the trunks, U^BC ANOVAR - Analysis of variance and covariance. Computing Centre, University of British Columbia, 1978. 69 pp. U^BC GENLIN - A general least square analysis of variance program. Computing Centre, University of British Columbia, 1977. 55 pp. ••Michigan Interactive Data Analysis System (Fox and Guire 1976). 23 using the proportion of opened galleries with pupal niches (Table XV), and yearly for the stumps (Table VII). This gave a measure of successful attacks (SA). SA = TA x PPN (2) The total number of pupal niches (TPN) in each host category and month were calculated by multiplying SA by the mean number of PN per successful gallery (MPNG) per month for the trunks (Table XV) and by the mean number of PN per successful gallery by year for the stumps. TPN = SA x MPNG (3) The estimated total number of brood (TB) were computed from total number of PN per month and the estimated number of brood/niche Table IX and XII) (BN) for the trunks and stumps. TB = TPN x BN (4) Equations 1-4 can be summarized as TB = (NA/PE) x PPN x MPNG x BN The estimated number of brood per month in each host category was then divided into emergence per month according to the pattern established in the galleries covered with petri dishes. These patterns were derived for three major periods: early summer attacks (Figs. 23, 24) for WH stumps and trunks, respectively; late summer 1978 attacks (Figs. 25-27) for WH trunks, Df stumps, and of trunks respectively; and May, 1979 attacks for a l l four host categories (Figs. 28-31). These results were then sum-marized to obtain the total numbers of brood produced by month for the total G_. sulcatus attacks. 24 3.0 RESULTS 3.1 HOST COLONIZATION BY Gnathotrichus sulcatus 3.1.1 Attacks on Each Host Category A total of 11,490 attacks by G_. sulcatus were pinned on the stumps, logs and downtrees during this study, 5212 in 1978 and 6278 in 197 9 (Table III). Of the total numbers pinned, 3222 attacks were on stumps with 1651 being recorded in 1978 and 1571 in 1979. Some 3561 attacks were pinned on trunks (logs plus downtrees) during 1978 and a further 4707 in 1979 for a total of 8268 attacks. There was a three fold difference i n the numbers of attacks recorded in the two study sites with 2980 and 8510 attacks being pinned on sites 1 and 2, respectively (Table III). In 1978, 4285 attacks were recorded on WH as compared to 927 on Df. In 1979, however, 3191 and 3087 attacks were recorded on WH and Df, respectively. On the stumps, 176 and 1475 attacks were recorded on site 1 and 2, respectively, during 1978, and in 1979, 630 and 941 attacks were pinned for these same two sites. One Df stump (Df 4) was not attacked in either year while two stumps that were not attacked in 1978 (Df 1 and WH 3) were successfully colonized in 1979. Stump DF 8 was not attacked in 1979, but sustained 115 attacks in 1978 (Table III). 3.1.2 Comparison of Attack Densities For stumps, higher densities of attack were recorded on WH than on Df. Higher densities of attacks were recorded in site 2 than in 25 TABLE III. Total number of Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps, logs (L) and downtrees (DT). UBC Research Forest, Maple Ridge, B.C., 1978/79 l 1978 1979 Trunk Pair Host Condition Stump Trunk Total Stump Trunk Total SITE 1 1 WH DT 70 78 148 216 10 226 Df DT 0 13 13 46 309 355 2 WH L 10 24 34 80 400 480 Df L 19 18 37 31 40 71 3 WH DT 0 84 84 102 12 114 Df DT 1 66 67 90 8 98 4 WH L 76 11 87 65 248 313 Df L 0 3 3 0 850 850 Total WH 156 197 353 463 670 1133 Df 20 100 120 167 1207 1374 Total attacks 176 297 473 630 1877 2507 SITE 2 5 WH DT 349 1085 1434 147 124 271 Df DT 53 127 180 130 574 704 6 WH L 281 478 759 208 776 984 Df L 15 151 166 179 668 847 7 WH DT 103 646 749 174 439 613 Df DT 81 79 160 18 132 150 8 WH L 478 512 990 85 105 190 Df L 115 186 376 0 12 12 Total WH 1211 2721 3932 614 1444 2058 Df 264 543 807 327 1386 1713 Total attacks 1475 3264 4739 941 2830 3771 XWH - Western Hemlock; Df - Douglas-fir 26 site 1. During the second year there were no differences in attack densities between the two sites, but again more attacks were pinned on WH than on Df stumps (Table IV). For logs and downtrees there were significant differences in attack densities by site, host and condition in 1978 (Table IV). Density of attack was approximtely six times higher on site 2 than on site 1. Significantly higher densities were recorded on WH than on Df trunks, and on downtrees than on logs during 1978. None of these differences was sustained i n 1979. On a bi-weekly basis, attack densities on trunks in 1978 showed a consistently and significantly higher attack density in site 2 from July 14 through November 2, while no significant differences were found for the bi-weekly periods ending in June 2 through June 30. WH was more heavily attacked than Df from July 14 through November 2, but again no differences were found from June 2 to June 30. Attack densities on downtrees were not significantly different from those on logs for any bi-weekly periods (Table V), but for final attack densities in 1978 there were significantly more attacks on downtrees than on logs (Table IV). For the period March 31 - May 29, 1979, higher attack densities were recorded on site 2 than on site 1. Those densities were not different from June 12, August 7, and for the periods ending August 21, and September 18 higher attack densities were recorded in site 1 (Table V). Significantly higher attack densities recorded on WH during 1978 were not sustained beyond March 31, 1979, there being no significant differences between hosts for the remainder of the year. 27 TABLE IV. Summary of densities of Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps, logs and downtrees. U.B.C. Research Forest, Maple Ridge, B.C. 1978-79. Factor 1.2 Total attacks/V Stumps" Logs and downtrees 1978 1979 1978 1979 SITE Site 1 Site 2 23.3b 162.3a 81.1a 97.8a 3.9b 26.2a 16.1a 19.6a HOST Western hemlock Douglas-fir 150.3a 35.3b 126.2a 52.8b 25.0a 5.1b 12.7a 18.0a CONDITION Log Downtree 13.6b 16.4a 22.4a 13.3a *Data transformed to X* = log 1 Q(X +1) prior to analysis. ^Means within each year/factor category followed by the same letter not significantly different, Neuman Keuls Test, P <0.05. 3AN0VA table in Appendix 1. 4AN0VA table in Appendix 2. TABLE V. Bi-weekly mean attack/m2 Gnathotrichus sulcatus on western hemlock and Douglas-fir logs and downtrees. UBC Research Forest, Maple Ridge, B.C. 1978/79. Mean number of attacks/m 1 9 7 8 J J A S O N 2 16 30 14 28 11 25 8 22 6 20 2 SITE 1 0.00a 0.04a 0.16a 0.18a 0.53a 0.50a 0.67a 0.81a 0.16a 0.42a 0.20a 0.09a 2 0.10a 0.47a 0.83b 1.50b 2.59b 4.59b 5.11b 4.49b 3.12b 2.19b 1.26b 0.31b HOST WH 0.10a 0.50a 0.95a 1.64a 2.65a 3.74a 5.00a 4.17a 2.75a 2.02a 1.04a 0.33a Df 0.00a 0.01a 0.04a 0.04b 0.47b 0.85b 0.78b 1.14b 0.53b 0.59b 0.42b 0.06b CONDITION L 0.00a 0.31a 0.71a 0.75a 1.01a 2.23a 2.48a 2.31a 1.75a 1.14a 0.71a 0.15a DT 0.10a 0.20a 0.28a 0.93a 2.12a 2.37a 3.30a 2.99a 1.53a 1.41a 0.80a 0.21a Means followed by the same letter, within each year/factor category not significantly different, Neuman Keuls Test, P <0.05. TABLE V. (Cont'd) — T 2 Mean number of attacks/m SITE HOST WH Df 1 9 7 9 - M •* . ,5 2 9 12 J 26 10 J 24 7 " 21 4 ' 18 2 16 30 31 13 CONDITION L DT 0.04a 0.03a 0.15a 0.51a 2.34a 1.41a 0.87a 0.75a 1.62a 1.71a 2.27b 1.14a 1.88b 0.74a 0.48a 9.29a 0.37b 0.14b 1.02b 2.58b 8.82b 2.95a 1.99a 0.96a 0.22a 0.29a 0.07a 0.07a 0.02a 0.00a 0.01a 0.00a 0.32a 0.13a 0.73a 2.10a 5.16a I.93a 1.45a 0.52a 0.89a 1.10a 1.49a 0.63a 0.91a 0.18a 0.08a 0.00a 0.09b 0.04a 0.45a 0.99a 6.00a 2.43a 1.40a 1.19a 0.95a 0.90a 0.85a 0.58a 1.00a 0.56a 0.41a 0.29a 0.14a 0.10a 0.90a 1.60a 5.82a 2.51a 1.68a 0.94a 1.36a 1.64 1.99a 0.95a 1.63a 0.55a 0.44a 0.29a 0.27a 0.07a 0.27a 1.49a 5.35a 1.84a 1.17a 0.77a 0.48a 0.37a 0.35a 0.26a 0.28a 0.19a 0.05a 0.00a 'Means followed by the same l e t t e r , within each year/factor category not s i g n i f i c a n t l y d i f f e r e n t , Neuman Keuls Test, P <0.05. 30 3.1.3 Seasonal Attack. Patterns The f i r s t two attacks were recorded on WH stump 8 on May 26, 1978, while for the logs and downtrees the f i r s t attacks were recorded on June 2. Numbers of attacks in 1978 increased steadily during June and highest numbers of attacks were recorded in late July and early August for the WH stumps (Fig. 6), and in late August and early September for WH logs and downtrees (Fig. 7). Peak attacks in Df stumps occurred on June 20 (Fig. 6), while the peak numbers of attacks on Df logs were recorded on September 8 (Fig. 7). Last attacks for 1978 were recorded for the period ending November 2 when the weekly maximum temperature was 13°C (Fig. 9). The f i r s t attacks in 1979 were recorded for the period ending March 25 when maximum temperature had reached 12°C. Highest numbers of attacks occurred on May 29 for WH and Df trunks on both sites (Figs. 6,7). This attack was sustained on site 1 trunks with a peak of attacks on WH in mid-August and on Df in mid-September. A f i r s t peak of attacks in the stumps in 1979 occurred on WH on both sites and on Df in site 1 on May 29. There was a general increase in attacks on Df stumps in site 2 which reached a maximum during August and September. The only other second peak recorded was that on August 21 for WH stumps in site 1. Last attacks were recorded on October 30 in 1979 when the maximum weekly temperature reached 13°C (Fig. 9). 3.1.4 Attack i n Relation to Moisture Content in Trunks and  Stumps Moisture content on WH stumps was approximately 100%, while for Df stumps i t was around 60% (Fig. 10). Higher moisture levels in WH 9 7 8 9 7 9 Figures 6, 7: Bi-weekly totals of pinned attacks of Gnathotrichus  sulcatus on western hemlock and Douglas-fir. UBC Research Forest. May-November/1978 and March-October / l 979. 6. Stumps. 7. Logs and downtrees. I 1 2 0 ^ 1 1 0 0 2 1 8 0 ~j 6 0 2 4 0 I 2 0 t222 8 : R A I N F A L L : : y 7 i t •.: :: • : ».. 9 : T E M P E R A T U R E — maximum — minimum M J J A S O 1 9 7 8 N D J F M A M J J A S 1 9 7 9 O N D Figures 8, 9: Climate data recorded at the gate of the UBC Research Forest from May 18, 1978 to December 31, 1979. Maple Ridge, B.C. 8. Weekly r a i n f a l l . 9. Minimum and maximum weekly temperature. Ill K <0 Z Ul o 101 STUMPS DOUGLAS -F IN HEMLOCK .1 - -Af\ A 5 2 0 1 | E H p l o w t o r y a t tack S u s t a i n * * a t tack 111 LOGS U AND DOWNTREES fe>T] 4 .2 HEMLOCK i, a. s, 7 |n) HEMLOCK 2,4, #,8 U DOUGLAS-FIR \ 3. % » OT DOUGLAS-FIR 1 4 , • , • ! . ] Ul a. ^1 \ N M M M J J A S O 1 9 7 8 1 9 7 9 Figures 10, 11: Mean moisture content of western hemlock, and Douglas-fir felled on May 8, 1978 In the UBC Research Forest for the periods May 1978 - November 1979. Initiation of sustained attacks by Gnathotrichus sulcatus indicated by solid arrows and fi r s t attack indicated by wavy arrows. Numbers refer to host pairs. Vertical bars represent standard errors of reported means. 10. Stumps; 11. Logs and downtrees. u> 34 were found in May, 1978, and these levels generally decreased to lowest values i n July and August. Df trunks had much lower moisture levels and varied less widely throughout the year (Fig. 11). Peak moisture measurements generally coincided with high precipitation levels (Fig. 8). Sustained attack (solid arrows Figs. 10,11) showed a wide variation between host categories. Sustained attacks occurred on six and two WH stumps and on one and six Df stumps in 1978 and 1979, respectively (Fig. 10). Five and three WH trunk and five and two Df trunk had sustained attacks in 1978 and 1979, respectively (Fig. 11). 3.1.5 Distribution Around Trunks Significantly higher percentages of attacks were recorded on the right, l e f t and bottom than on the top quadrants in WH 1 m long logs. No differences between these percentages could be demonstrated for Df. maximum number of attacks/m^ on WH and Df were 135.4 and 160.8 attacks/m^ respectively (Table VI). 3.1.6 Evaluation of Pinning Efficiency Pinning efficiency for the disks that were dissected was 70.0% (257/367) and 66.1% (214/324) respectively for WH and Df logs. On WH stumps 81.7% (428/524) of the attacks were pinned, while on Df stumps 63.0% (199/316) were marked. For the 1 m long logs, 77.6% (274/353) of the attacks on WH logs and 60.3% (140/232) of those on Df logs were pinned. Overall pinning efficiency for WH and Df combined was 70.8% (414/558). The lowest pinning efficiency for both hosts was recorded on the bottom quadrant 35 TABLE VI. Distribution of Gnathotrichus sulcatus attacks by quadrants on lm sections of Douglas-fir and western hemlock logs. UBC Research Forest, Maple Ridge, B.C. May-November, 1978 and March-October, 1979. N = 8 for each species. 1 2 Percent attack Attacks/m 2 Mean + S.D. Range Quadrant —  WH Df WH Df Top 13.3 + 9.4a 15.6 + 9.5a 0 - 80.8 0 - 62.0 Right 24.7+ 6.9b 28.1 + 17.4a 12.8- 92.4 0-160.8 Left 28.2 + 4.8b 28.2 + 22.2a 17.2 - 101.2 0 - 87.6 Bottom 33.6 + 10.1b 28.0 + 9.5a 8.4 - 135.4 4.8 - 78.0 1Means followed by the same letter within each host not significantly different. 2AN0VA table in Appendix 3. 36 of logs where frass f e l l directly to the ground and so failed to accumulate and make the attack v i s i b l e , and i t was often d i f f i c u l t to observe these trunk surfaces directly. 3.2 HOST COLONIZATION BY ASSOCIATED SCOLYTID SPECIES 3.2.1 Dendroctonus pseudotsugae, the Douglas-fir Beetle Two peaks of attacks were recorded. The f i r s t after f e l l i n g in May and the second in September. In both cases, logs sustained more attacks than downtrees (Fig. 12). After the f i r s t peak, the numbers of attacks decreased and reached 4 attacks on downtrees and 16 attacks on logs on August 18. There was a strong second peak of attacks on September 15. There were no attacks after October 13. No significant differences in total attacks between sites or between logs and downtrees were demonstrated (Appendix 4). 3.2.2 Scolytus unispinosus, the Douglas-fir Engraver Attacks of this bark beetle started on June 9, and reached a peak of 302 on June 23. After this date, attacks decreased and no attacks were recorded after September 15 (Fig. 13). Significantly more beetles attacked downtrees than logs in 1978 (Appendix 4). A total of 469 attacks were recorded on site 1 and 59 on site 2, and 98% of the attacks recorded on site 1 occurred on downtree No. 3. On site 2, 48 were on downtree No. 7 and 9 on downtree No. 5. Attacks were concentrated on the top quadrant of attacked material. 37 o < Q Ui 160 J 140. 120. 100 80 60 40 20 1 2 : D w i d r o c t o n u 8 pseudotsugaw OC UJ 03 2 3 310 300. 1 l t f f 100J 90 80 70J 60 SO 40 30 20 10 1 3 : Scolytua uniaotnoaua JZL n n n • DOWNTREES H LOGS 26 M 23 J 21 J 18 A 15 S 13 O 1 9 7 8 Figures 12, 13: Seasonal attack pattern of two bark beetles on Douglas-fir at the UBC Research Forest in 1978. Maple Ridge, B.C. 12. Dendroctonus pseudotsugae. 13. Scolytus unispinosus. 38 3.2.3 Scolytus tsugae (Swaine) the Hemlock Engraver Attacks started with two attacks on June 9 and reached a peak on July 7 for both downtrees and logs, after which they declined until no attacks were recorded on September 29 and beyond (Fig. 14). Mean number and range of attacks was 18.2 (3-41) and 70.0 (10-120) on logs and downtrees, respectively, which was significantly different only at 7% level (Appendix 4). For the bi-weekly periods, the only differences appeared for the periods ending on June 23 and August 4, when s i g n i f i -cantly more attacks were pinned on site 1 than on site 2 and on August 4, more S^. tsugae attacks were pinned on site 2 than site 1. Total attacks recorded on site 1 and site 2 were 186 on 187 respectively. 3.2.4 Pseudohylesinus tsugae Swaine Attacks started on June 9, 1978, with a high level of attacks from June 23 to August 4, decreasing until no attacks were recorded on November 9. When considering downtrees alone, the peak of attacks was earlier, on June 23, while for the logs the peak was reached only on August 18 (Fig. 15). Mean number and range of attacks was 60.2 (14-161) and 104.5 (6-324) for logs and downtrees respectively. No significant differences between the number of attacks in the two trunk conditions were shown for any bi-weekly period (Appendix 4). S i g n i f i -cantly more attacks were pinned on site 2 than on site 1 for the periods ending July 23, August 4 and September 1. 39 1 4 : Sco ly tus tsugae CO < I-< O 111 u. O oc UJ CD • DOWNTREES £ LOGS M 1 9 7 8 Figures 14, 15 Seasonal attack pattern of two bark beetles on western hemlock at the UBC Research Forest in 1978. Maple Ridge, B.C. 14, Scolytus tsugae; 15, Pseudohylesinus tsugae. 40 3.2.5 Trypodendron lineatum (Olivier), the Striped Ambrosia Beetle No attacks were recorded in 1978 on either WH or Df. In 1979 the i n i t i a l attack started on April 24, and reached a peak on May 1 for downtrees and logs. Attack rates decreased steadily in both downtrees and logs until practically no attacks were pinned from July through the end of the year (Fig. 16). No significant differences were established between sites or hosts in the analysis of variance for total attacks per square meter (Appendix 5). Mean number and range of attacks/m^ were 4.2 (0-17.7) and 12.4 (0-43.5) for WH and Df, respectively. 3.3 ESTABLISHMENT SUCCESS OF Gnathotrichus sulcatus The percentages of successful establishment of G_. sulcatus over two years were 82.0% and 35.5% for WH and Df stumps, respectively. Attack success was generally higher in the second year. In 1978 and 1979 the percentages of successful attacks recorded were 77.1% and 82.0% for WH stumps and 18.5% and 61.1% for Df stumps, respectively (Table VII). Out of 49 and 59 galleries covered on WH and Df trunks, respectively, 73.5% and 25.4% were successful over two years. On a yearly basis, success rates of 72.9% and 75.0% for WH trunks and 17.6% and 85.7% for Df trunks were achieved in 1978 and 1979, respectively (Table VII). Similar rates of successful establishment were achieved in the two trunk conditions for WH and Df. For logs and downtrees the percentages of success were 75.0% (9/12) and 70.8% (17/24) in WH and 20.0% (6/30) and 31.0% (9/29) in Df respectively. Such results reflect 41 1000. en : <•> 5 0 0 . 4 0 0 . o 3 0 0 . CL U. O 2 0 0 . DC U l GO Jg 100 . 13 A T r y p o d e n d r o n l i n e a t u m 15 M ^ W E S T E R N HEMLOCK • D O U G L A S - F I R J 2 12 J 10 J 7 A 1 9 7 9 Figure 16: Total bi-weekly numbers of Trypodendron lineatum attacks pinned on western hemlock and Douglas-fir in 1979 at the UBC Research Forest, Maple Ridge, B.C. TABLE VII. Number and success of Gnathotrichus sulcatus attacks monitored with petri dishes on western hemlock stumps and trunks and Douglas-fir stumps and trunks in 1978 and 1979, UBC Research Forest, Maple Ridge, B.C. Western hemlock Douglas-fir Stumps Trunks Stumps Trunks number Succ.1 Unsucc. Total Succ.1 Unsucc. Total Succ.1 Unsucc. Total Succ.1 Unsucc. Total 1 5 0 5 1 5 6 0 0 0 5 2 7 2 7 1 8 4 3 7 3 6 9 1 9 10 3 8 1 9 5 1 6 2 5 7 1 7 8 4 5 0 5 6 0 6 0 0 0 0 2 2 5 4 2 6 5 1 6 5 5 10 3 5 8 6 5 1 6 5 1 6 5 0 5 5 5 10 7 2 4 6 6 0 6 0 7 7 0 6 6 8 5 0 5 4 2 6 1 6 7 0 8 8 Year Totals (percentage success ful) 1978 27 8 35 (77.1) 27 10 37 (72.9) 5 22 27 (18.5) 9 44 51 (17.6) 1979 14 1 15 (93.2) 9 3 12 (75.0) 11 7 18 (61.1) 6 1 7 (85.7) Total 41 9 50 (82.0) 36 13 49 (73.5) 16 29 45 (35.5) 15 42 59 (25.4) 1Defined as successful after producing frass for two weeks and producing four adults. 43 the steady sustained attacks on most WH during 1978, while for most Df sustained attacks were achieved only during the second year (Figs. 10, 11). 3.4 FRASS PRODUCTION Significantly higher amounts of frass (FR) were recorded from galleries on WH than from those on Df i n a l l host categories. S i g n i f i -cantly more frass was produced from galleries on downtrees than from those on logs. Generally higher frass production occurred on stumps than on trunks (Table VIII). Frass production was higher immediately following gallery i n i t a -tion. The duration and length of frass production varied by attack date. Attacks in the summer of 1978 produced lower amounts of frass per week but the production was sustained throughout the spring and the summer of 1979 (Figs. 17-20). Attacks in May 1979 produced larger quantities of frass for 6-10 weeks after which production rates declined rapidly (Fig. 17-20). Frass was always white at the beginning of gallery construction, but after 5 weeks, on the average, i t was discolored by the presence of black fungal hyphae (Fig. 3). It became darker every week until no wood particles were clearly visible in the frass. 3.5 BROOD PRODUCTION 3.5.1 Stumps in the Field Larger numbers of brood (BR) were produced on WH than on DF (Table IX). 44 TABLE VIII. Mean total frass production from 147 galleries of Gnathotrichus sulcatus monitored on western hemlock and Douglas-fir stumps and trunks in the UBC Research Forest, Maple Ridge, B.C. 1978-79. Stumps 1,3 No. galleries mg/frass monitored gallery Trunks 1,3 No. galleries monitored mg/frass gallery SITE 1 2 HOST WH Df TRUNK CONDITION Logs Downtrees 30 233.36a 37 130.26a 40 209.49b 40 142.48a 40 313.84a 40 202.90a 30 94.22b 37 65.00b 37 122.12b 40 140.17a XData transformed to X' = log 1 Q(X + 1) prior to analysis. 2Significance levels indicated results of ANOVA using GENLIN, means followed by the same letter within each pair not significantly different, P <0.05. 3AN0VA table in Appendix 6. 25 L 0» J 20 UJ 15 10 UJ or UJ o. 0) oc •Jt- 3 5 U- 3 0 O . 2 5 20 UJ 15 10 Z < 5 UJ 17: W E S T E R N H E M L O C K l o g s and d o w n t r e e s B-V W E S T E R N H E M L O C K s t u m p s Figures 17, 18: A S O N M A M J J A S O N 1 9 7 8 1 9 7 9 Mean weekly frass production of galleries initiated at various times during 1978 and 1979 by Gnathotrichus sulcatus on western hemlock, at the UBC Research Forest. 17. Logs and downtrees, A - 5 attacks from June 2-June 30; B - 27 attacks from July 14-September 29; C - 5 attacks from May 1-June 5; D - 4 galleries on May 15. 18. Stumps, A - 26 attacks from June 2-July 7; B - 6 galleries from May 1-May 22; C - 6 galleries from May 22-June 5. O) E 20 -* UJ UJ 15 . £ DC 10 . UJ Q. 5 . (/> CO < oc u. i u. 20 o 15 X o UJ 10 z 5 < UJ 5 9: D O U G L A S - F I R l o g s a n d d o w n t r e e s 1 v.. / '* J x ;• •-s t u m p s A / I ••f *L.T-\ j j A S O N M A M J J A 1 9 7 8 1 9 7 9 Figures 19, 20: Mean weekly frass production of galleries initiated at various times during 1978 and 1979 by Gnathotrichus sulcatus on Douglas-fir, at the UBC Research Forest. 19. Logs and downtrees, A - 8 attacks from Sept. 1-Sept. 22; B - 6 attacks from May 1-May 22. 20. Stumps, A - 5 attacks from July 28-Sept. 1; B - 11 attacks from May 15-June 5, 1979. 47 TABLE IX. Brood production from 147 galleries of Gnathotrichus sulcatus monitored on western hemlock and Douglas-fir stumps and trunks in the UBC Research Forest, Maple Ridge, B.C. 1978-79. Stumps 1,3 Trunks 1,3 No. galleries Mean no. ^ monitored brood/gallery No. galleries Mean no. 2 monitored brood/gallery SITE 30 40 11.6a 6.4a 37 40 8.1a 5.4a HOST WH Df 40 30 13.7a 1.9b 40 37 10.7a 2.3b TRUNK CONDITION Logs Downtrees 37 40 8.2a 5.4a iData transformed to X' = log 1 0(X + 1) prior to analysis. 2significance levels indicated results of ANOVA using GENLIN, means followed by the same letter within each main effect not significantly different, P <0.05. 3AN0VA table in Appendix 6. 48 The peak number of attacks by sulcatus i n 197 9 on site 1 (Figs. 6, 7) coincided with peak brood production (Fig. 21). On site 2, the f i r s t peak of attacks coincided with high brood production but the f a l l peak of brood production was not matched by a peak of attacks at the same time (Figs. 6,7 and 22). Attacks covered in early summer, June and early July 1978 (Figs. 23, 24) produced a major contribution of beetles to the spring 1979 f l i g h t . In the case of WH stumps, brood started to emerge in the same year in low numbers from September to November. By June 5, 1979, however, 50% of the brood had emerged (Fig. 24). Brood continued to emerge from this material throughout the summer and 90% had emerged by late August. Late summer attacks on Df stumps (July through October) in 1978, produced low numbers of brood throughout the following summer with 50% of the brood being produced by August 7 and 90% by late August (Fig. 27) 3.5.2 Trunks in the Field Brood production was significantly higher on WH than on Df and there were neither differences in the numbers of brood produced on each site nor between brood production in logs or downtrees (Table IX). In addition to higher brood production, WH trunks also had higher densities of attacks (Table IX). Attacks covered in early and late summer produced no brood in the same year on either WH or Df. Few beetles emerged in the spring of 1979, and most of the brood was produced from the end of June to the end of August 1979 (Figs. 23,25,26). For the early summer, 50% of the 180 160 140. 120. 100 80 J 60 Q 40 O O 20 OC 03 L. o 1 2 0 OC Ui 1 0 o CO 3 80 z 60 40 21: Sitel g * , g I IP 9 «P 22: Site 2 18 13 15 12 10 4 4 3 M A M J J A S O 1 9 7 9 30 Figures 21, 22: Total brood production galleries monitored in from June 197 8 to June 21. Site 1, (N = 67). from Gnathotrichus sulcatus the UBC Research Forest 1980, Maple Ridge, B.C. 22. Site 2 (N = 80). > cc U J < o oc UJ a Q o o oc 00 u. O oc UJ CO Z Z < UJ 5 2-0 1-5 1 0 0 5 2 0 1-5 1 0 0 5 231 W E S T E R N H E M L O C K logs and downtrees At tack date : June 1 6 - 3 0 , 1978 5 GALLERIES |BP50-5-22 i i i 1 IBP90«902 241 W E S T E R N H E M L O C K s t u m p s A t tack da te : June 2 - Ju ly 7, 1978 26 GALLERIES BP 50 •7 43 JL BP90=13-60 • Pi ihh, S O N M 1 9 7 8 M J J A S 1 9 7 9 Figures 23, 24: Mean weekly brood production from Gnathotrichus sulcatus g a l l e r i e s established on western hemlock i n June - early July (early summer). UBC Research Forest, Maple Ridge, B.C. B P 5 0 and BPgQ are the f i f t y and ninety percentiles for cumulative brood production. 23. WH trunks. 24. WH stumps. 51 25:WESTERN HEMLOCK logs and downtrees B P s o. 7. 7 8 ATTACK DATE - July 14-Sept. 29 1978 27 GALLERIES 1 -5 . RY 1 0 . UJ _J 0 - 5 . < O cc UJ i 0 . OD 1 -5 . O cc 1 0 . CO u. 0 - 5 . o cc UJ m D Z 1-5. Z < UJ 1 0 5 0 5 I I I 1 BP00- 9-89 H l A L i ATTACK DATE - Sept. 1-22 1978 8 GALLERIES BPso. 6 73 BP90 = 10 09 27: DOUGLAS-FIR stumps ATTACK DATE - July 28 - Sept. 1 1978 5 GALLERIES J L _ J BP 50.38 BP 90.6-4 MLi M M 1 J 9 A 9 Figures 25-27: Mean weekly brood production from Gnathotrichus sulcatus galleries established on WH and Df from late July to October, 1978, UBC Research Forest, Maple Ridge, B.C. 25. WH stumps. 26. Df trunks. 27. Df stumps. 52 brood was out by July 24 (Fig. 23), while for the late summer attacks 50% was out by August 7 for WH (Fig. 25) and August 14 for Df (Fig. 26). For May 1979 attacks, 50% of the brood emerged by September 4 for WH (Figs. 28, 29) and August 28 for Df (Figs. 30, 31), indicating that, as for the stumps, the spring attacks made a major contribution of brood to the peak populations in the f a l l . 3.5.3 Logs i n Laboratory Successful galleries were established on three WH logs (3, 4 and 5 galleries) and one Df log (4 galleries) (Table X). On both hosts, there was steady brood production over time, with the last brood emerging on day 192 (Fig. 32). Higher numbers of brood were produced on Df than on WH (Fig. 32). There were no significant differences between these two hosts for either length of gallery or depth of penetration into the logs. The four successful galleries on Df produced 56 males and 57 females, with 3 males and 2 females being the f i r s t to emerge from the galleries. For WH, 12 out of 31 galleries were successful in the three logs. These galleries produced 46 males and 48 females, with 8 males and 5 females being the f i r s t to emerge. One gallery produced 2 males on the f i r s t day. For WH, the mean time to emergence the f i r s t brood was 76.6 + 13.9 days after the female was put into the gallery. For Df the mean time to f i r s t emergence was 90.0 + 10.42 days. 53 3 >• 2 DC UJ 1 ml - J < o cc 3 u 0 . 2 Q O 1 o OC 03 u . O 3 OC 2 UJ CD 1 Z Z < 3 HI 2 2 [28i WESTERN HEMLOCK logs and downtrees ATTACK DATE - May1-June5 1979 I 5 GALLERIES * * * n A • 2» WESTERN HEMLOCK stumps ATTACK DATE - May 8-June 12 1979 I 13 GALLERIES A BPsorfO-08 BP9CMS-32 30> DOUGLAS-FIR logs and downtrees ATTACK DATE - May 1-22 1979 I 6 GALLERIES BPM -877 I BP90J4-35 JL 31: DOUGLAS-FIR stumps ATTACK DATE - May 15-June 5 1979 11 GALLERIES BPM>*-97 BP9O-13-01 M J J A 1 9 7 Figures 28-31: Mean weekly brood production from Gnathotrichus sulcatus galleries established on WH and on Dr in May-June, 1979. UBC Research Forest,, Maple Ridge, B.C. 28. WH logs and downtrees. 29. WH stumps. 30. Df logs and downtrees. 31. Df stumps. 54 TABLE X. Total numbers of brood produced and sex ratio from Gnathotrichus sulcatus galleries established on western hemlock and Douglas-fir under laboratory conditions, UBC 1979. Sex ratio 1 Gallery WH 1 WH 2 WH3 Df 1 WH Df M F M F M F M F M:F M:F A 8 1 2 6 3 2 19 19 1.44 1.00 B 2 5 10 6 3 3 17 14 1.07 1.21 C 2 5 5 3 3 6 11 9 0.71 1.22 D - - 4 3 2 4 6 15 0.86 0.4 E - - - - 2 4 - - 0.5 -Totals 12 11 21 18 13 19 53 57 0.96 0.93 X 4.0 3.7 5.2 4.5 2.6 3.8 13.2 14.2 S.E.(M) 1.0 1.3 1.7 0.9 2.4 0.7 3.0 2.0 Results of the analysis of variance in Appendix 7. - No brood were produced. 3 0 . 2 8 . > CC UJ 26 . _ J 2 4 . _ l < o 22 . cc Ul 2 0 . OL 18 . o o 16 . o cc 14 . CO u. 12 . o 1 0 . DC UJ' CO 8 UM 6 z 4 z < 2 UJ z 70 8 0 9 0 100 110 120 130 140 150 160 D A Y S A F T E R F E M A L E W A S P L A C E D I N G A L L E R Y 180 190 200 210 Figure 32: Cumulative mean Gnathotrichus sulcatus brood production in western hemlock and Douglas-fir logs in laboratory conditions, Faculty of Forestry, UBC. Numbers in parentheses indicate number of successful galleries in each host log. 56 Oviposition occurred in groups on the lateral galleries near to the main gallery but some galleries had pupal niches in the main branch (Fig. 33). These side galleries continued for several centimeters without any pupal niches. On WH, the pupal niches were concentrated on 14.3% of the total length of galleries, the comparable figure for Df was 28.5%. 3.5.4 Sex of Emerging Beetles The sex ratio of beetles reared in the laboratory was not different from 1:1 in both hosts either for total or f i r s t emerging beetles (Table X). Of the 103 galleries classified as successful in the f i e l d work, 34 were on WH trunks, 38 on WH stumps, 15 on Df trunks and 16 on Df stumps. First emerging beetles had a sex ratio of 1.16 males : 1 female, not significantly different from a 1:1 sex ratio (Chi-square Test P <0.05). Sex ratio of total brood was 1.09 males : 1 female (Table XI). 3.6 GALLERY DISSECTIONS 3.6.1 Stumps There were no significant differences in the number of egg niches (EN) per gallery either by site or host (Table XII). For the development profile (Fig. 34) of G. sulcatus the egg niches represented 3.0% and 26.8% of the total number of niches (EN and pupal niches) for WH and Df respectively. The number of pupal niches (PN) was significantly higher for WH than for Df and no differences between the two sites were found. Flo„rP 33- Schematic plan of Gnathotrichus sulcatus gallery in western hemlock two Figure 33. Schematic^ N o t e t h e pT-^ET5f egg niches (EN), pupal niches (PN), dead beetles (D), and live beetles (A). Ul 58 TABLE XI. Sex ratios of Gnathotrichus sulcatus emerging from western hemlock and Douglas-fir trunks and stumps felled May 1978 in the UBC Research Forest, Maple Ridge, B.C. First week* Total emergence Host Number of galleries M F Sex ratio M:F M F Sex ratio M:F WH stumps 37 36 23 1.56 354 309 1.15 Df stumps 16 15 13 1.15 115 99 1.16 WH trunks 34 32 35 0.91 264 268 0.99 Df trunks 15 14 16 1.14 106 93 1.14 iGalleries often produced more than one beetle during the f i r s t week. 59 TABLE XII. Mean egg niches (EN), pupal niches (PN) length of gallery (LG) and depth of penetration (DP) of Gnathotrichus sulcatus galleries on western hemlock and Douglas-fiTTt^pT and trunks. UBC Research Forest, Maple Ridge,' B.C. monitored niches STUMPS Sites: 1 30 0.6a 14.0a 31.2a 5.2a 2 40 1.1a 8.8a 28.4a 5.3a 40 0.5a 16.2a 38.2a 7.1a 30 1.5a 4.1b 18.1b 2.8b Hosts: WH Df TRUNKS Sites: 1 2 Hosts: WH Df Condition: Logs Downtrees 37 1.4a 10.3a 28.2a 4.4b 40 1.3a 7.2a 29.4a 5.3a 40 1.6a 14.1a 41.6a 7.0a 37 1.1a 2.9b 15.0b 2.5b 37 1.2a 10.0a 27.2a 4.7b 40 1.4a 7.5a 30.3a 5.0a l D a t a transformed to X' - log 1 0(X + 1) prior to analysis. 3AN0VA table in Appendix 8 and 9. M E A N P E R G A L L E R Y IN) 09 C 1 CD O J - P -• • cn 3 r r (B C « 3 3 •a cn ro era ft, (TO 3 a. 3 —u r r T 3" C <T> 3 -?r cn -3 C » -a 3 M a. I-1 o 3 3 n a 3" i - h (0 cn 93 r r 3 C a. 3 "3 o* CO o S> o 3 o. a. •a r r n it o C &. 3 c ?r o CO r r • O 3 a 0 8 O o i-h 70 O <T> 3 CO W f t r r Cu 3" l-f O o r r •y i t TS n o 3" i-t C (D CO CO r r CO » c o w •a r r i — 1 C fD CO so era CD Q. M cra rt) (Tl « r-t H -fD * CO • o 3 m c/> H m 33 z m O o o o c o r" > I (0 >•» c 3 0) o to a " i + 3 9 CD (0 (0 C •o (0 o 0) a i + 3 »* " i (D (D 0) m >f E G G N I C H E S 00 a> j 09 _L_ O J rO 0> 00 L A R V A L - P U P A L N I C H E S B R O O D -I 3 II o II o 1 II w o m 1 II W 09 61 Fewer brood developed than the number of pupal niches present, indicat-ing larval and pupal mortality. The difference ranged from 15.4 to 53.7% of the total number of pupal niches on WH and Df, where mean number of pupal niches found was 16.2 and 4.1 per gallery on each of the two hosts, respectively (Table XII). No parasites, diseases or predators were captured, but in several cases, dead pupal or callow adults were found in the pupal niches. Galleries were longer and deeper on WH than on Df (Table XII). Mean length of galleries (LG) and depth of penetration (DP) for WH were 38.2 and 7.1 cm while for Df they were 18.1 and 2.8 cm respectively. 3.6.2 Trunks No differences In number of egg niches could be demonstrated either for site, host or condition. Mean number of egg niches were 1.6 and 1.1 for WH and Df respectively (Table XII). There was a significantly higher number of pupal niches in WH than on Df, with no significant differences related to site or condition (Table XII). The number of pupal niches that failed to develop a brood represented 24.1% and 20.7% of total pupal niches in WH and Df respectively (Fig. 34). The mean number of pupal niches found was 14.1 and 2.9 for WH and DF, respectively. Longer and deeper galleries were found in WH than in Df trunks. The galleries had mean lengths of 41.6 cm and 15.0 cm, and depths of 7.0 cm and 2.5 cm for WH and Df, respectively (Table XII). 62 3.6.3 Relationships Between Brood, Frass and Length of  Galleries Significant relationships were found between brood and frass, brood and gallery length and frass and gallery length for both WH and Df trunks and stumps (Table XIII). Examination of the gradients of the regression lines suggest some consistent relationships between the measured variables. Highest number of brood per cm of gallery was recorded for WH trunks and stumps, one beetle for each 2.38 and 2.86 cm of galleries respectively. Additionally, one beetle was recorded for every 12.35 and 22.73 mg of frass for WH trunks and stumps, respectively. For the same species, i t was found that 1 mg of frass was equivalent to 0.21 and 0.12 cm of gallery for stumps and trunks, respectively. Galleries in Df had much lower productivity on a per cm gallery basis than those in WH, with one beetle for each 16.95 and 27.03 mg of frass for trunks and stumps, respectively. Additionally one beetle was found for each 5.7 and 8.3 cm of gallery for trunks and stumps, respec-tively. For frass and length of galleries i t was found that 1 mg of frass was equivalent to each 0.22 cm and 0.23 cm of galleries for Df trunks and stumps, respectively. 3.6.4 Aspects of Gallery Construction by Gnathotrichus sulcatus  at Different Periods of Time After Felling. A total of 269 galleries, approximately equally distributed between logs and downtrees, at site 1 and site 2 and by host species, were dissected. TABLE XIII. Relationships between brood production, frass production and length of galleries of Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps and trunks in the UBC Research Forest, Maple Ridge, B.C. Host category Brood and frass Brood and length Frass and length WH stumps Br = 0.044 Fr;** 1 r = 0.72 Br = 0.35 LG;** r = 0.51 Fr = 8.06 LG**; r = 0.71 WH trunks Br - -5.812 + 0.081 Fr**; r = 0.81 Br = -6.82 + 0.42 LG**; r = 0.68 Fr = 4.86 LG**; r = 0.78 Df stumps Br = -1.49 + 0.037 Fr**; r = 0.80 Br = 0.12 LG**; r = 0.55 Fr - 4.39 LG*; r = 0.59 Df trunks Br = -1.54 + 0.059 Fr**; r = 0.84 Br = 0.175 LG**; r = 0.49 Fr = 4.47 LG**; r = 0.80 iSignificance levels of the regression indicated: 2When the constant is not present in the equation, through zero. * = P <0.05; ** = P <0.01. i t was not significant and the regression was put On 64 There were no differences for site, host, condition or month in numbers of egg niches when analysing by host. No differences were found between sites, conditions or months for WH, and the only difference for Df was the higher number of egg niches in logs than in downtrees. Mean number of egg niches per month showed a high variability in each host (Fig. 35). Significantly higher numbers of pupal niches were recorded on WH than on Df, 7.21 and 4.87, respectively. Numbers of pupal niches ranged from 3.12 to 9.55 per gallery per month (Fig. 36). When considering each host separately no difference appeared for site or month, but significantly more pupal niches were recorded for logs than for downtrees on WH (Table XIV). For Df alone, no differences appeared for month and condition, but significantly more pupal niches were recorded for site 1 than site 2 (Table XIV). On WH, 141 galleries were opened with 83 showing pupal niches, while 58 had no pupal niches. For Df, 128 galleries were dissected, with pupal niches in 55 galleries, but none in 73 others. The number of galleries with pupal niches was higher in WH than on Df from the beginning (Table XV). For WH the lowest proportion of galleries with pupal niches was 38.2% i n June 1978. A l l other months except August 1978 had showed more galleries with pupal niches than without them. For Df, galleries initiated in June, 1978, had no pupal niches. Only in the summer of 1978 were there more galleries with pupal niches than without them (Table XV). 35: EGG NICHES '"WESTERN HEMLOCK 36 PUPAL NICHES DOUGLAS- FIR Figures 35, 36: Mean number of egg and pupal niches of Gnathotrichus sulcatus on western hemlock and Douglas-fir trunks from June 1978 to October 1979. UBC Research Forest. 35. Egg niches. 36. Pupal niches. TABLE XIV. Egg niches, pupal niches, length of galleries and depth of penetration of Gnathotrichus sulcatus on western hemlock and Douglas-fir trunks over 13 months. UBC Research Forest, Maple Ridge, B.C. 1978-79. Frequency WH/Df Total 1 Mean gallery parameters EN WH" Df PN WH Df LG (cm) WH" Df DP (cm) WH Df SITE 1 2 58/62 83/66 120 149 0.98a 0.88a 1.00a 0.79a 7.43a 7.06a 6.69a 24.94b 17.22a 4.85b 2.54a 3.15b 28.60a 14.54a 5.68a 2.37a HOST2 WH Df 141 128 0.92a 0.89a 7.21a 4.87b 27.09a 5.34a 15.84b 2.45b CONDITION Logs 73/63 136 1.07a 1.40a 9.56a 3.93a 27.68a Downtrees 68/65 133 0.76a 0.40b 4.68b 5.75a 26.67a 16.Ola 5.16a 2.32a 15.67a 5.53a 2.58a ^Means followed by the same letter within each host, site, condition and/or month combination not significantly different. 2Data on host analysis were taken from the combined western hemlock plus Douglas-fir analysis. ^Results of the analysis of variance in Appendix 10. *Re suits of the analysis of variance in Appendix 11. TABLE XV. Number of Gnathotrichus sulcatus galleries with and without pupal niches, and mean number of pupal niches on western hemlock and Douglas-fir trunks over 13 months. UBC Research Forest. 1978/79. Western hemlock Douglas-fir Month 1 2 With No Total Mean Mean PN PN PN 1 2 1 2 With No Total Mean Mean PN 1 2 PN PN _ 1978 June July August September October 1979 March April May June July August September October Total 5 8 6 7 6 6 6 9 8 9 7 3 3 83 8 7 8 6 4 2 2 5 7 4 3 1 1 58 53 81 97 88 100 52 84 111 171 66 73 19 24 1019 10.6 10.1 16.2 12.6 16.7 8.7 14.0 12.3 21.4 7.3 10.4 6.3 8.0 4.1 0 4 0 0 0 5.4 4 4 24 6.0 3.0 6.9 1 11 4 4.0 0.3 6.8 6 9 75 12.5 5.0 10.0 5 9 28 5.6 2.0 6.5 3 3 41 13.7 6.8 10.5 3 5 44 14.7 5.5 7.9 4 10 73 18.2 5.2 11.4 7 7 120 17.1 8.6 5.1 8 4 128 16.0 10.7 7.3 7 2 69 9.9 7.7 4.7 4 • 3 16 4.0 2.3 6.0 3 2 46 15.3 9.2 55 73 668 ^ o t a l number of pupal niches divided by the 2Total number of pupal niches divided by the niches (PN)). number of galleries with pupal niches. total number of galleries (with and without pupal 68 Galleries were significantly longer on WH than on Df but no differences appeared for site or condition. For WH and Df combined there were shorter galleries produced by June, 1978 attacks than in any other month, but no difference could be demonstrated in each host alone (Table XIV). Considering WH alone, the only significant value was longer galleries on site 1 than on site 2, but no differences appeared for condition or month. For Df alone no differences appeared either for site, condition or month. Except for July, 1979, WH had consistently longer galleries than Df (Fig. 37). There were significantly deeper galleries in WH than in Df. Mean depth of penetration ranged from 3.56 to 4.49 cm among months (Fig. 38) when considering WH alone, no differences appeared either for condition or month, but deeper galleries were recorded on site 2 than on site 1. The depth of penetration on WH ranged from 3.91 to 6.01 cm. For Df alone, no differences appeared either for site, condition or month. As for this last factor, depth of penetration ranged from 1.67 to 3.10 cm deep; mean gallery depth exceeded 3 cm only i n July 1978 (Fig. 38). 3.7 PRIMARY ATTRACTANTS TESTS 3.7.1 Western Hemlock Stimuli The sulcatol-baited traps were removed at the end of the f i r s t week because they were catching the majority of the beetles (2721/2747) and thus probably reduced catches in those traps baited with host materials. Results taken from the last two weeks were not used in the Figures 37, 38: Gnathotrichus sulcatus gallery parameters on western hemlock and Douglas-fir trunks from June 1978 to October 1979. UBC Research Forest. 37. Length of galleries. 38. Depth of penetration. 70 analysis because only 3 beetles were caught in the fourth week and none in the last week. There were significant differences between catches by week (Appendix 12) with catches in the third week being lower than in the f i r s t two weeks. The total catch in three weeks to traps baited with the host stimuli was 56 beetles, of which 33 were caught in sapwood-baited traps, significantly more for both sexes (P <0.01) than in any other treatment (excluding sulcatol-baited traps) (Table XVI). Sulcatol-baited trap results were not included in the analysis because the relatively large catches in these treatments would introduce variation into the analysis which could prevent differences between host stimuli being established. The interactions week x block and block x treatment were signi-ficant (Appendix 12). Such interactions reflect block v a r i a b i l i t y (P <0.01), with low catches in block 5 during week 3 contributing greatly to the significant week by block interaction and the low catch of treatment 4 (heartwood) in block 3 contributing to the significant block by treatment interaction. The number of beetles caught per block varied widely. For example, 23 beetles (13 males:10 females) were captured over three weeks in block 3 while only 5 males were caught in block 1. These results may have been a consequence of uneven distribution of infested host material throughout the forest. Only 3 and 4 beetles were caught, respectively, in the f i r s t and the second experiment in 1979. 71 TABLE XVI. Numbers of Gnathotrichus sulcatus caught in response to the odours of bark, sapwood and heartwood of western hemlock and in bark, phloem, sapwood and heartwood of Douglas-fir. Point Roberts, WH. June-July, 1978. Stimuli Number of beetles captured Male Female Total Western hemlock experiment2»-> Mean weight stimulus per trap (g) Percent of moisture content Beginning End Blank control 2b lb 3b — — Bark and phloem 8b Ob 8b 477.0 59.5 18.1 Sapwood 20a 13a 33a 417.5 81.5 19.2 Heartwood 8b 4b 12b 390.0 42.8 18.8 Douglas-fir experiment3 ,5 Blank control Ob Ob Ob - -Bark lb Ob lb 403.3 33.1 Phloem 19a 8a 27 a 673.6 - 95.3 Sapwood 2b lb 3b 422.0 65.6 Heartwood Ob Ob Ob 361.8 28.7 •^Totals followed by the same letter within each column in each experiment, not significantly different, Neuman-Keul's Test, P <0.05. 2Total catch of five blocks run over three weeks. 3Catch in one week, in three blocks. ^Moisture content of fresh materials. 5AN0VA table in Appendix 12. 72 3.7.2 Douglas-fir Stimuli Only two beetles were caught in block number 1 and none in block number 2. Since such numbers were too low to provide meaningful data on treatment preferences, these two blocks were excluded and the analyses of variance were done using data from the three remaining blocks. There were no significant differences between numbers of beetles captured in these blocks. The numbers of beetles caught by the phloem stimuli were significantly higher for males, females and total beetles (Table XVI) than for any other treatment. Out of 22 males caught, 19 were to this stimulus, and of the 9 females, 8 were caught in the same treatment (Table XVI). Again, significantly more males than females were caught (Appendix 12), consistent with the role of the male as the pioneer beetle. 3.8 TOTAL BROOD EMERGENCE PATTERN The overall emergence pattern from the 16 spring-felled trunks and their stumps showed a few brood emerging in September and October, 1978 from WH stumps attacked the previous June. Most of the brood emerged in 1979 when a steady increase in brood emergence was recorded from July, reached a peak in August and decreased through October (Fig. 39). Attacks from March to June, 1979 produced most of their brood in the f a l l of the same year, but 5,523 beetles were produced in 1980, along with 25,477 beetles from June to October, 1979 attacks (Table XVII). 73 3 0 ~ 2 5 t o o o w 2 0 f — - T o t a l a t t a c k s ( x 2 ) B r o o d f r o m a t t a c k s i n i t i a t e d : • J u n e 1 9 7 8 • J u l y - O c t o b e r 1 9 7 8 M a r c h - J u n e 1 9 7 9 0 N M A M J 1 9 8 0 Fieure 39: Estimates of total monthly attacks and brood production during 1978 through 1980 for western hemlock and Douglas-fir host material felled in May 8, 1978. UBC Research Forest, Maple Ridge, B.C. TABIE XVII. Brood production p r o f i l e from the estimated 16,049 Gnathotrichus sulcatus attacks on western hemlock and Douglas-fir stumps and trunks from June, 1978 to June, 1980. UBC Research Forest, Maple Ridge, B.C. WH Stumps WH Trunks Df Stumps Df Trunks Date of attack Early 1978 Late 1978 Early 1979 Estimate of total attacks 545 1,122 777 Estimate of Total Brood Production (by month) 1978 1979 1980 s 1, ,236 0 0 250 0 N 0 0 M 177 57 A 150 0 H 1, , 8 6 3 « 2 57 J 478 920 J 1, ,382 4, ,259* A 774 3, ,782 S 277 2, ,158 0 119 545 Total 5 ,706 11, ,778 M A M J Total 0 0 0 195 4,395 5,029* 1,293 10,912 50 0 1,450* 144 1,644 Early 1978 206 0 0 0 13 0 92 66 241* 184 38 13 Late 1979 3,537 0 0 0 90 0 90 1,463 6,772* 6,014 3,431 867 Early 1978 2,056 0 0 0 186 4,699* 4,135 1,879 Early Late Early 1978 1978 1979 Early Late Early 1978 1978 1979 129 M z s 323 328 10 1,054 2,579 647 18,727 10,899 153 382 2,130 0 2,401 12,498 0 0 124 0 0 111 0 0 143 0 0 62 0 0 27 0 0 143 0 0 2,752* 0 0 97* 0 0 143 0 0 248 0 0 29 0 0 0 3,186 264 429 Total 12,666 0 0 0 0 236 0 0 0 0 250 0 0 0 0 0 0 0 337 11 0 0 0 0 161 22 0 0 53 0 2,177 33 0 0 53 0 3,013 109 248 0 766 286 14,444 142* 955* 0 818* 6,590* 28,353* 65 775 0 555 4,915 21,378 0 152 0 156 707 5,731 76,080 428 232 4,442* 421 5,523 Total brood from June-October 1979 attacks 3 Total brood Total attacks Brood per parent 1978 1979 c o m b i n e d 8,710 38,750 2,983 5.2 8.1 6.5 3,005 36,464 6,466 2.6 3.1 2.8 1 A d d i t i o n a l 3,383 a t t a c k s J u n e - O c t o b e r , 1979. 2 * = p e a k uf b r o o d p r o d u c t i o n f o r e a c h a t t a c k p e r i o d . 3 No p e t r i d i s h e s s e t o n a t t . i r k s fr.™ l l . i s , . e r i . . . l , e s t i m a l e s b.ise.l »i i d i s k d t s s u . - t i o n s . 3,278 6,207 1,236 0.6 3.6 2.5 10,484 25,812 5,364 1.1 2.7 2.4 25,477 107,080 16,049 3.3 4^ 75 The number of beetles representing attacks, i.e. corrected number of attacks pinned x 2, was higher than brood produced throughout 1978 and up to June 1979. From July to October 1979, the brood produc-tion was many times higher than the number of beetles representing attacks (Fig. 39). From 1978 to 1980 a total of 107,080 brood were produced from an estimated total 16,049 attacks, which gives an average of 3.3 beetles produced for every parent. In a l l four host categories more beetles per parent were produced from 1979 attacks than from 1978 attacks. 76 4.0 DISCUSSION 4.1 HOST SELECTION AND COLONIZATION BY Gnathotrichus sulcatus An estimated total of 16,049 attacks were recorded on the test materials after correcting the total pinned attacks with the pinning efficiency values obtained from debarked trunk and stump samples. Attempts to establish field-collected brood (J. sulcatus on hosts of varying ages in the laboratory were largely unsuccessful. It was hypothesized that the beetles may have required flight exercise before entering a host selection/establishment behavior pattern as has been suggested for T. lineatum (Graham, 1959) and ID. pseudotsugae (Atkins, 1966). The more rapid colonization of site 2 in 1978 suggested that the site was closer to the flight path or to the source of dispersing (J. sulcatus. The effect of host odour is a close range one and may be more of an arrestant than an attractant as was indicated by the low catch of G_. sulcatus to host materials (approximately 1% of total catch) as compared to sulcatol (Section 3.7). In addition, primary attractants are also important to females as demonstrated by the catch of 8 and 13 females, compared with 19 and 20 males to WH sapwood and Df phloem respectively. This showed that the females were also attracted by host volatiles although to a lesser degree than the pioneer beetles. The more lik e l y way for G_. sulcatus to find suitable host material i s : f i r s t , random flying, which disperses the beetles and 77 gives them flight exercise; second, close range orientation to host material releasing primary attractants; and third, establishment of secondary attraction in suitable hosts which wi l l attract beetles of both sexes to ensure f u l l host u t i l i z a t i o n . Sustained attacks occurred sooner on WH than on Df which suggested that attacking beetles were able to establish galleries, produce pheromone, and maintain secondary attraction more easily on WH stumps and trunks. The more rapid colonization of WH by G_. sulcatus may be related to the fact that the source of primary attraction Is located in the sapwood, the area of gallery formation, whereas in Df the primary attractants apparently emanate primarily from the phloem (Table XVI). In the latter case the beetle can f l y to an attractive but, perhaps unsuitable host, thus delaying the establishment of secondary attraction in Df. This hypothesis is supported by the high failure rate of attempts to establish "successful" galleries in Df trunks and the high number of short unproductive galleries (Table XV). Better host suitability of WH was demonstrated by higher i n i t i a l success rates (Table VII) and higher brood production (Table IX) than on Df. Preference shown by (3. sulcatus for Df over WH reported by McLean and Borden (1977a) may have been a result of synergist effects between Df host attractants and the sulcatol used to bait the logs. Previously, G_. sulcatus was shown to attack and breed more readily in WH green lumber than in Df lumber (McLean and Borden 1975b). The single peak of attack in 1978 (Figs. 6,7) indicated that the trees were probably felled too late for colonization by any minor 78 spring f l i g h t , while the number of attacks increased to a peak in late summer. This is consistent with previous results on spring-felled host material in this area (McLean and Borden 1977a). These authors hypothesized that their host materials were unsuitable for attack in a f a l l flight period. My data have demonstrated that there are as many attacks during the second year as in the f i r s t and that the lack of a f a l l flight was more related to the lack of suitable host material during the May attack period that would have produced brood by the f a l l than any host suitability factor per se. Higher numbers of attacks on downtrees supported the i n i t i a l hypothesis that downtrees would be attacked more than the logs, but disagreed with Johnson (1964) who reported higher attacks on logs. Later work by the same researcher failed to demonstrate any difference in attack rate on logs and downtrees (Johnson and Zing 1969). In the forest, downtrees created by the action of wind, snow or ice are the natural hosts for ambrosia beetles. These downtrees retain their crown most of the time during the f i r s t months. As an aid to increased survival, G. sulcatus should be able to locate and colonize such trees. The response to downtrees in this study supported the possible explana-tion that passive liberation of host attractants was increased with water lost by the needles. This theory could be tested by placing unbaited sticky traps near the top and near the butt of downtrees. Moisture content in downtrees dropped more rapidly than that in logs, especially for WH (Fig. 11). The continual loss of water may have facilitated the transport and evaporation of primary attractants. 79 The stumps, logs and downtrees in this experiment probably never lost water to a degree at which dryness would affect survival of brood. McLean and Borden (1977b) showed that G. sulcatus readily attacked boards with a moisture content of 62.3% but did not attack boards drier than 46%. A lack of moisture has also been suggested as a factor limiting attack by T. lineatum (Kinghorn 1956). Attack densities on logs and downtrees did not reach the mean level of 182.3/m^  recorded on stumps. This suggested that logs and downtrees may have had a lower nutritional value than stumps for fungal growth, that they became non-attractive more quickly, or that phago-stimulant production ceased at an earlier stage. For example, WH downtree 5 sustained 53.7 and 6.1 attacks/m2 i n 1978 and 1979, respectively, while WH log 8 had 61.4 and 12.6 attacks/m2 for the same periods. This indicates that logs and downtrees might have a similar population density when attacks decrease or cease. This decrease of attacks agreed with past findings that G^. sulcatus does not continue attacking material already heavily infested (Prebble and Graham 1957; McLean and Borden 1977a). This was further supported in this study by an apparent switch from one host to another that occurred after June 26, 1979 when fewer beetles attacked the heavily infested WH stumps and trunks. Borden and Stokkink (1973) suggested that a switch off or masking mechanism might occur in G_. sulcatus as has been reported for T. lineatum (Nijholt 1973). The high number of attacks may have resulted in high levels of pheromone production that might 80 have repelled other beetles. Alternatively, microbial degradation of bark and sapwood may have rendered these stumps and trunks unsuitable. The lower attack rate in the top quadrant for WH may have reflected dryness, high temperatures or higher light intensity (McLean and Borden 1977a). The last two are particularly important for this beetle, because i t is a crepuscular species for which flight i s inhibited by high light intensity (Daterman et a l . 1965); thus i t w i l l l i k e l y prefer shaded areas. Other ambrosia beetles, such as J_. lineatum have reduced attack rates on exposed logs (Dyer 1963a). No differences between quadrants could be demonstrated in Df because this species had only two trunks directly exposed to the sun, while WH had five trunks, similarly exposed, in the samples taken. The lowest pinning efficiency was recorded on the bottom sector because of greater d i f f i c u l t y experienced in observing fresh attacks on surfaces where frass f e l l directly from the trunk. The use of a chisel to remove part of the bark and frass around galleries increased pinning efficiency as often more than one attack was found under a pile of frass. This was especially true on the stumps, where frass accumula-tion per unit area was higher, because of the greater density of attacks. Another advantage was that the beetles were more positively identified. 81 4.2 HOST SELECTION AND COLONIZATION BY ASSOCIATED SCOLYTID BEETLES There appeared to be no cross attraction or repulsion between (J. sulcatus and P_. tsugae and S. tsugae on WH and (J. sulcatus and D. pseudotsugae and j>. unispinosus on Df. In WH, .P. tsugae attacks during the f i r s t eight weeks were mostly found in downtrees, but later this trend disappeared. The flying period extended into mid-September, which was similar to that recorded by McLean and Borden (1977a). This extended flight period is typical of tsugae which overwinters at any larval stage, pupates, and emerges in July (Bright and Stark 1973). Only one attack was found in a WH stump, which does not confirm the preference for stumps reported by Bright (1976). This species always started i t s galleries under bark scales, similar to the pattern reported by McLean and Borden (1977a). About 80% of the attacks by j>. tsugae occurred on WH down-trees, almost significantly more than on logs (P <0.07). The low level of attacks in one of the downtrees contributed much variation. Attacks were recorded from June to mid-September which indicated a longer flight period than that reported by Bright (1976), who reported that flight by this species usually extended from May to late July. When the trees were felled, the flight of JJ. pseudotsugae was probably under way because on the f i r s t bi-weekly period, May 26, peak numbers of attacks were recorded on logs (160 attacks) and on downtrees (66 attacks), followed by a second minor peak on September 15. The f i r s t peak, probably represented insects which overwintered as adults and the second peak those, which overwintered as larvae or re-attacks 82 by emerging adults after establishing a gallery (Walters 1956; Skovsgaard 1968; Furniss and Carolin 1977). Most of the attacks by S. unispinosus, the other bark beetle in Df, were recorded in late spring and early summer, which agrees with Bright (1976). Overall, more attacks occurred on downtrees than on logs. A heavy concentration of attacks occurred in a single downtree. The major difference between the two most attacked downtrees and the other two was that the f i r s t pair was in f u l l sunlight, which may have increased liberation of chemical attractants from these downtrees and attracted beetles in higher numbers to them. This agrees with Deyrup (1981), who reported that this bark beetle prefers trunks under f u l l sunlight while i t s ecological twin P. nebulosus prefers shaded trunks. The other ambrosia beetle recorded in the study was T_. lineatum. No attacks were recorded in 1978 because the logs were felled on May 8 and this beetle attacks only logs that have aged (Prebble and Graham, 1957). The quick build-up of attacks in 1979 was characteristic of T_. lineatum (Fig. 16), when the temperature reached 15.5°C in the spring, which is the threshold for flight (Chapman and Nijholt 1980), and the beetles l e f t hibernation sites in the forest duff (Hadorn 1933). Even with higher attack rates in site 2 and on Douglas-fir no •differences could be demonstrated either for site or host. This was probably due to the high variation that existed within the same host. For example the attacks for Douglas-fir were concentrated on Douglas-f i r No. 5, a downtree, an aggregation probably mediated by secondary attraction. 83 4.3 ESTABLISHMENT AND GALLERY PARAMETERS The most visible difference between the two years was that attacks in 1979 had a quicker establishment with higher amounts of frass production in the f i r s t ten weeks. Such higher frass production and success ratio in 1979 indicated a better suitability of the host or a stronger beetle after a long maturation feeding period through the winter, thus allowing £. sulcatus to make galleries more rapidly. After this period of ten weeks, frass production gradually decreased during the remainder of the year indicating that the main phase of gallery construction was completed. Low levels of frass from August to November might be mainly frass produced by larva enlarging their niches. Frass was steadily produced from successful galleries which indicated that when such a gallery was established the parents did not leave i t . On the other hand, many galleries, especially those in Df stumps and trunks covered with petri dishes, stopped producing frass in three weeks to two months after attack when one or both of the beetles abandoned the gallery. This demonstrated that the establishment phase proposed by Borden (1974) takes several weeks before successful establishment occurs. The presence of black fecal material in frass from successful galleries was recorded five weeks after commencement of attacks indicating that at this time the beetle had already started eating black hyphae.. In a l l four host categories, frass production started to drop at the end of July for 1979 attacks when brood started to emerge. It 84 i s hypothesized that a maturation feeding period (Batra 1963) i s necessary and with brood walking through the gallery system further boring by parents i s inhibited. In many cases i t was observed that one of the parents blocked the gallery entrance and prevented any brood from re-entering the gallery once they were in the petri dish. The production of very low levels of frass, mostly black material in 1979, from 1978 attacks, indicated that at this point the parents were only removing fecal material and frass produced by larvae excavating their niches. A l l galleries opened were clear of frass showing that these materials are routinely removed. The only host category that produced similar amounts of frass per gallery for the attacks in both years was WH stumps. Thus, the other three host categories were not completely suitable for beetle reproduction during the f i r s t months after f e l l i n g . This is confirmed by the fact that WH stumps showed the highest number of attacks per square meter and also the highest number of brood per gallery. One i s a consequence of the other, because i f the i n i t i a l beetles are successful they w i l l attract others. Approximately 300 old attacks identified by colour-coded pins, which had stopped producing frass, produced fresh white frass between August and October 1978, indicating reactivation. Such reactivation had to be done by beetles from outside, rather than brood from the same galleries, because no covered unsuccessful attack produced any fresh material later in the season. A consequence of such reactivation i s 85 that should the beetle abandon a previously pinned gallery and then reattack i t would be counted twice. Many galleries, opened after frass production had ceased, had empty egg niches. There are four possible explanations for the presence of empty egg niches: lack of oviposition following niche - construction, i n f e r t i l i t y , a direct host effect preventing egg hatch, and cannibalism. There was no evidence to support the third hypo-thesis, since there were no differences in number of egg niches per gallery between hosts, even though WH supported much more successful brood production. If cannibalism by brood was a factor, i t should have been proportional to numbers of brood and no such correlation could be obtained. The most probable explanation is that the female made the egg niche, but no egg was laid in i t . Since the potential for egg production should be similar in a l l females within WH or Df, lower numbers of pupal niches in the latter, probably indicated that the female may have obtained better nutrients from WH woodsap than from Df during gallery construction, resulting in greater production of eggs in WH. Alternatively, i t could have indicated poor establishment of the fungi, and that the females lay eggs only after feeding on their fungi. The greater depth of G_. sulcatus galleries in WH was a conse-quence of much deeper sapwood and lack of fungitoxic heartwood in this tree species. In some of the WH trunks, galleries extended almost to the pith. The longer galleries not only produced more brood but also caused greater degrade to WH than to Df. In the latter species 86 galleries were confined to the sapwood and they terminated at the well differentiated heartwood. Significantly negative constants on the brood and frass relationships equations (Table XIII) indicated some loss of frass since petri dishes were put on one week after attack and covered one week after installation. A specific gravity was calculated for stumps and trunks on both hosts using the relationships between frass and volume of galleries. For WH stumps and trunks specific gravities of 0.62 and 0.37 were found, respectively. Since in this species the early wood zone usually occupies two-thirds or more of the ring, and transition from early to late wood is more or less gradual (Panshin and Zeeuw 1980), specific gravity of the early wood should be smaller but closer to the specific gravity for the whole log, which varies from 0.429 to 0.487 (Jessome 1977). Specific gravity of early wood is about 0.25 (Wilson 1968). The specific gravity for trunks of 0.37 is intermediate between these values as would be expected for galleries which crossed annual rings. As for the stumps an unusually high specific gravity of 0.62 was found, which could not be explained. Even though the number of pupal niches in stumps was the highest overall, equal to 16.25 per gallery as compared to 14.10 for the trunks, this alone does not explain the high specific gravity found in the stumps. There may have been a higher contribution of fungal biomass in the WH galleries. Specific gravity for Df stumps and trunks were both 0.34. This i s consistent with Parker et al. (1976) who reported specific gravity 87 values ranging from 0.313 to 0.346 for early wood which contrasts with the average value of 0.20 found by Wilson (1968). The early wood zone in Douglas-fir is usually several times wider than the band of late wood (Panshin and Zeeuw, 1980) and most of the G_. sulcatus galleries w i l l be constructed in the early wood, which is reflected by the specific gravity values. These results again suggested some contribu-tion to frass weight by the beetle digestive residues. 4.4 ASPECTS OF GALLERY CONSTRUCTION OF Gnathotrichus sulcatus  IN TRUNKS FOR DIFFERENT PERIODS AFTER FELLING Few s t a t i s t i c a l l y significant differences could be demonstrated in the data because of high va r i a b i l i t y in the relatively few numbers of observations but some trends are worth mentioning. Mean numbers of pupal niches for WH and Df were 7.21 and 4.81, respectively, as compared to 14.10 and 2.89, found for the brood production work. The latter value for WH i s probably overestimated, since petri dishes continued to be installed until successful galleries had been established whereas the f i r s t reported means above are averaged over a l l dissected galleries. Much lower numbers of PN per gallery in WH and Df logs were found by McLean and Borden (1977a) than those registered here, because they opened galleries before brood production had been completed. Mortality within galleries due to natural enemies is probably not important for G. sulcatus since the mean PN per gallery were similar in covered galleries (successful + unsuccessful) on WH and 88 in uncovered galleries. For Df, even though many petri dishes were set up (Table VII), they were in the most part put on during 1978 when this host was not completely suitable as demonstrated by the low number of pupal niches found for galleries initiated in 1978 (Table XV). Probably for this reason, mean numbers of PN were lower in the galleries covered with petri dishes, than the mean PN found when galleries were opened on a monthly basis. For WH, no numerical trend by month was apparent, but in Df i t was clear that lower numbers of pupal and egg niches and shorter galleries were recorded from 1978 attacks (Table XV) which demonstrated that spring-felled logs of this latter host are not suitable for colonization during the f i r s t year. This trend confirmed host unsuitability as indicated by a lower number of attacks during 1978 in Df (Table III). No differences were demonstrated for months within either host. No galleries were opened until August 1980, by which time a l l attacking beetles including those of October 1979 should have completed gallery construction and brood production. Beetles that attacked WH and Df in October 1979, 17 months after f e l l i n g , were able to successfully complete the cycle and cause damage similar to that of earlier attacks (Table XV). About 70% of total length of galleries had no pupal niches and most pupal niches were adjacent to the radial entrance of the gallery. It i s possible that the whole gallery i s built to enable more fungus cultivation for the adult and larval feeding. If this were the case, 89 the larvae of G. sulcatus would need to move in the gallery system in a similar manner as those of Xyleborus celsus (Gagne and Kearby 1979). The difference between these two species is that X. celsus larvae feed communally in a single chamber, while G_. sulcatus have individual "cradles" to which they would need to return after any feeding. The a b i l i t y of G. sulcatus to colonize logs and downtrees in the second year after f e l l i n g as observed during this study is in agreement with other authors (Mather 1935; Prebble and Graham 1957). Dyer and Chapman (1965) further commented that (3. sulcatus preferred Df felled in the previous year meaning that Df needs considerable aging before being completely suitable for G_. sulcatus attacks. 4.5 RELATIONSHIP BETWEEN BROOD EMERGENCE PATTERN AND ATTACK DATA The estimated total of 107,080 brood produced over a two year period in these 16 stumps and trunks probably represented more than one generation, because attacks in the f a l l 1978 and during 1979 could have been by brood produced by the earliest attacks (Fig. 39). G. sulcatus attacking WH stumps in the second year had an eight fold generation increase but on Df stumps the brood would only replace the original parents. Thus, WH stumps provided a better environment for brood production. Similar results were found for logs and downtrees. Brood production in the laboratory showed a steady pattern of emergence throughout 200 days with no peaks similar to those seen in the f i e l d . A similar pattern was reported by Hosking (1972) for X. saxeseni in New Zealand, in laboratory conditions. This result 90 suggested that brood emergence in the f i e l d must be co-ordinated by an interaction between environmental factors, such as temperature and light intensity. It has been demonstrated by Rudinsky and Schneider (1969) that few G. sulcatus f l y i f light intensities exceed 21,520 lux. No meaningful degree-day calculations could be made using the lab and f i e l d results. It may have been that the laboratory tempera-ture used was above the optimum for development of the brood or the symbiotic fungi. Numbers of brood per parent in the f i e l d ranged from 0.6 to 8.1, depending on the host and time of attack (Table XVII), as compared to an overall 4.5 brood per parent found by Dyer (1963b) for T_. lineatum on stumps, logs, chunks and tops of Df in a clearcut. Probably the shade and the partial shade conditions in which trees were felled in this experiment might have contributed to the high numbers of success-ful attacks and brood production. More pupal niches per gallery were found for J_. lineatum by Gibson et^ al. (1958) on shaded than on exposed sites in logs of Df. Also, more brood per gallery were found for _T. lineatum on shaded than on sunlit sides of logs, trunks, tops and stumps of Df (Dyer 1963b). Further studies could elucidate the brood productivity pattern in stumps and logs in shaded and exposed clearcut sites, as well as by season of the year in which logs were cut. In this study, brood production was very low in late summer and f a l l , 1978, when the highest numbers of attacks were recorded. These attacks probably represented activity by migrating beetles that completed their development elsewhere. The brood production and attack 91 peaks in May 1979 showed a greater contribution of brood from WH stumps than from any other host material. This suggests that brood were re-attacking material in the same general area. Peak numbers of attacks in 1979, for a l l host materials (Figs. 6,7), were higher than that in 1978 due to brood production from galleries established the previous year plus improved host susceptibility. Brood would attack in the same area provided they had sufficient flight exercise. For each time of attack a pattern of brood emergence was observed. When a l l emergence was combined i t was shown that G_. sulcatus had a f i r s t peak of emergence in the spring, dropping in June, reaching a peak in August and decreasing again through October. Under f i e l d conditions of this study i t was demonstrated that June 1978 attacks in WH stumps produced a peak of brood in May 1979. These beetles may have been the attacking beetles pinned at this time. The high brood population from August to September could be accounted for mainly by late summer, 1978 and spring, 1979 attacks and beetles from attacks up to June, 1979 contri-buted to the July to October, 1979 emerging population, with additional production in the following spring. From July to October 1979, total brood production exceeded the number of pinned attacks (Fig. 39) which indicated a net production of G. sulcatus, which would have dispersed to other areas. The emigration of beetles, as indicated by lower attack rate during summer and spring, 1979, indicated a decrease in host suitability. Most material is apparently unsuitable for heavy attack 15 months beyond a May f e l l i n g date. 1 92 The data (Table XI) do not support the hypothesis that the non-pioneer beetle i s the f i r s t to emerge to ensure outbreeding (Borden 1981) since 97 males and 87 females were the f i r s t to emerge from a l l successful galleries in the f i e l d and this is not different from 1:1 sex ratio (Chi-square test). On the other hand, i t is d i f f i c u l t to visualize inbreeding in a G_. sulcatus population since there are thousands of beetles and they a l l do not attack the same log year after year and have to move out, thus favouring outbreeding. Slightly more males than females were recorded (83 9 males : 769 females), which could allow for higher mortality among the pioneer beetles since the female can f l y directly to a male occupied gallery after emergence, decreasing flying time and thus decreasing mortality among them. The overall brood production pattern during 1979 (Fig. 39), is very similar to the pattern of pinned attacks in 1978 (Figs. 6,7) and that reported by McLean and Borden (1977a). This shows that in a forest setting where a (3. sulcatus population can be maintained genera-tion after generation, the expected pattern w i l l be high population from July to September. The peak of attacks recorded in May, 1979 was minor compared with the brood emergence numbers from July to September. Numbers of attacks probably did not build up to a real high peak during the summer and early f a l l of 1979, because the host materials were already heavily colonized. For this reason the relative importance of the recorded May 1979 peak was overestimated because i t was compared with 1978 peaks of attack when populations in the area were lower than 93 In 1979. This demonstrated that i f suitable host material was available in an area from year to year, population build up of (J. sulcatus could be considerable. Bi-modality in numbers of attacks as recorded by McLean and Borden (1975b; 1979) in a sawmill happened because the beetles caught are from outside populations (McLean 1980b). I suggest that the b i -modality can be explained as follows: f i r s t , high numbers of flying (3. sulcatus in May come from logs that were cut and attacked in the previous summer; second, logs processed during the summer period were cut during the winter, w i l l have few attacks and produce few brood; fi n a l l y , logs processed in August and September were cut during winter, yarded and attacked in the spring and w i l l produce a high number of beetles in late summer and early f a l l . Additional evidence to support this theory was shown by McLean and Borden (1977b) when no second flight peak was recorded during a mid-summer mi l l shutdown and no logs were coming from the forest. By the time the industry started up again in October i t was too cold for beetles to emerge from logs attacked in the spring. The results of this study, wherein 107,080 brood were estimated to have been produced from 16 stump/trunk combinations, showed that 42% were produced by stumps and 58% by logs. In the forest this would mean that stumps could sustain a relatively high number of G_. sulcatus and further, that field-infested logs transported to dryland sorting areas and mills could contain many thousands of beetles. Approximately 25 infested logs would be needed to produce the 50,000 beetles captured in 94 a pheromone-baited trap survey of the Shawnigan and Port McNeil dryland sorting areas (McLean 1980a). How many beetles survive the water transport phase to the mill? A 1 week per month survey of the Chemainus Sawmill in 1974 caught 3,098 (J. sulcatus (McLean and Borden 1975b) while efforts to suppress the population, captured 5,796 in 1975 (McLean and Borden 1977b) and 42,907 in 1976 (McLean and Borden 1979). 4.6 IMPLICATIONS OF THE RESULTS FOR THE B.C. FOREST INDUSTRY High attack rates and brood production on WH stumps and trunks poses a potential major threat to the forest industry, since WH makes up 40% of the volume processed in coastal British Columbia (B.C.M.F. 1980). The rapid build up of (3. sulcatus in WH, the extensive gallery construction and resultant serious degrade of this tree species emphasizes the need for rapid removal of WH from forest settings (hot logging) and rapid inventory turnover in storage and processing areas. The potential for (2. sulcatus build up in felled and bucked areas is immense as demonstrated by the production of 107,080 brood from 16,049 attacks. This number of attacks is probably an over-estimation since a beetle which attacked an unsuitable stump or trunk and re-emerged could have attacked another stump or trunk, and thus been counted twice. Another way to look at the reproductive capacity of £. sulcatus is that a single one meter high partially shaded WH stump can produce 3,574.3 + 2,085.1 beetles. This suggests that a f i e l d of 200 WH stumps could, produce 714,860 + 417,020 beetles. If a trap log or trap bundle tactic is to be used to capture (J. sulcatus i n dryland sorting areas then i t is recommended that WH 95 logs should be used as they received a mean of 18.2 sustained attacks/m 2, while Df received only 0.9 sustained attacks/m2 during the f i r s t year. Two types of WH exist, one with furrowed bark and the other with flaky bark (Walters et a l . 1960). The furrowed bark type has a purple phelloderm and should be used for trap logs, because my - observations have indicated that i t is preferred by ( 3 . sulcatus over the flaky barked type. Data developed in this study suggest that four month old WH logs can be used as trap logs at two different times of the year. First, they should be set out in March to remove G_. sulcatus from the spring f l i g h t , but most of the attacks w i l l probably be _T. lineatum. These logs would need to be removed by the end of June to prevent emergence of T_. lineatum brood. A second set of trap logs should be set out at the beginning of July to receive attacks from the heavy summer and early f a l l G_. sulcatus flight and then removed for processing before the following February. The increased use of dryland sorting and storage areas as opposed to water sorting as well as the planned increase in the use of barges for log transport w i l l almost certainly increase beetle popula-tion. Protection gained by water storage and transport as well as mortality caused by submergence of logs which drowned any brood present (Dyer and Chapman 1962, McLean and Borden 1977b) w i l l be greatly reduced. Additionally, most of the logs in a barge are shaded, which w i l l lead to higher survival of G_. sulcatus. 96 5.0 SUMMARY Aspects of G. sulcatus biology and host colonization c l a r i f i e d in this study. For spring-felled logs and downtrees as well as their stumps: 1. Western hemlock was the preferred host over Douglas-fir in the f e l l i n g year as indicated by numbers of attacks and success of establishment. No differences between trunks of the two species appeared in the second year, but again western hemlock stumps were preferred over Douglas-fir stumps. 2. More attacks were pinned on downtrees than on logs in the f i r s t year, but no differences were demonstrated in the second year. 3. Highest numbers of attacks were recorded during July, August and September of 1978 and on May, 1979 for a l l host categories. 4. More brood, as well as longer and deeper galleries were found in WH than on Df. 5. Peak of total brood emergence was recorded in August and September, 1979. Peaks of brood emergence from each host category varied by time of attack. 6. Western hemlock sapwood and Douglas-fir phloem were shown to be the most attractive host tissues and suggested starting points for further elaboration of primary attractants for (2. sulcatus. 97 Implications related to management of G!. sulcatus suggest that: 7. It is very important to trap the minor spring flight of G_. sulcatus, because spring attacks in this study produced 63.4% of total brood production in August and September. 8. Western hemlock, rather than Df, should be used in trap bundles, as i t i s more readily colonized. 9. Stumps produced 42% of the total brood production in this study showing that they are major sources of continuing infestation in the f i e l d . 10. Logs, especially western hemlock, should be removed as soon as possible after f e l l i n g (hot logging) to minimize degrade losses and the build up of G. sulcatus populations. 98 6.0 LITERATURE CITED Anonymous, 1972. Damage by ambrosia (pinhole borer) beetles. Technical Note No. 55. Princes Risborough Laboratory. Aylesbury, Bucks. England. 8 pp. Atkins, M.D. 1966. Laboratory studies on the behavior of the Douglas-fir beetle, Dendroctonus pseudotsugae Hopkins. Can. Entomol. 98:953-991. Bain, J. 1974. Overseas wood- and bark-boring insects intercepted at New Zealand ports. N.Z. For. Serv. Tech. Pap., No. 61. 24 pp. Batra, L.R. 1963. Ecology of ambrosia fungi and their dissemination by beetles. Trans. Kans. Acad. Sci. 66:213-236. Blackman, M.W. 1931. A revisioned study of the genus Gnathotrichus Eichhoff in North America. J. Wash. Acad. Sci. 21:265-276. Borden, J.H. 1974. Aggregation pheromones in the Scolytidae. In: M.C. Birch (ed.). Pheromones. North Holland, Amsterdam. Borden, J.H. 1981. Aggregation pheromones. _In: J.B. Mitton and K.B. Sturgeon (Eds.) (E.D.S.). Bark Beetles in North American Conifers, Ecology and Evolution. Univ. Texas Press, Austin. (In press). Borden, J.H. and E. Stokkink. 1973. Laboratory investigation of secondary attraction in Gnathotrichus sulcatus (Coleoptera: Scolytidae). Can. J. Zool. 51_: 469-473. 99 Borden, J.H., B.S. Lindgren, and L. Chong. 1980a. Ethanol and ot-pinene as synergists for the aggregation pheromones of two Gnathotrichus species. Can. J. For. Res. 10:303-305. Borden, J.H., L. Chong, K.N. Slessor, A.E. Oehlschlager, and H.D. Pierce, Jr. 1980b. Allelochemic activity of aggregation pheromones between three sympatric species of ambrosia beetles. Can. Ent. (In press). Borden, J.H., J.R. Handley, J.A. McLean, R.M. Silverstein, L. Chong, K.N. Slessor, B.D. Johnston, and H.R. Schuler. 1980c. Enantiomer-based specificity in pheromone communication by two sympatric Gnathotrichus species (Coleoptera:Scolytidae). J. Chem. Ecol. 6^ :445-456. Bright, Jr., D.E. 1976. The bark beetles of Canada and Alaska. Can. Dep. Agric. 241 pp. British Columbia Forest Service. 1980. Annual Report of the Ministry of Forests. Ministry of Forests. 60 pp. Byrne, K.J., A.A. Swigar, R.M. Silverstein, J.H. Borden, and E. Stokkink. 1974. Sulcatol: population aggregation pheromone in the scolytid beetle, Gnathotrichus sulcatus. J. Insect. Physiol. 20:1895-1900. Cade, S.C. 1970. The host selection behaviour of Gnathotrichus  sulcatus LeConte (Coleoptera:Scolytidae). Ph.D. Thesis, University of Washington, Seattle. 100 Cade, S.C., B.F. Hrutfiord, and R.I. Gara. 1970. Identification of a primary attractant for Gnathotrichus sulcatus isolated from western hemlock logs. J. Econ. Ent. 63:1014-1015. Chapman, J.A. 1962. Field studies on attack flight and log selection by the ambrosia beetle Trypodendron lineatum (Oliv.) (Coleoptera:Scolytidae). Can. Ent. 94:74-92. Chapman, J.A. 1963. Field selection of different odors by scolytid beetles. Can. Ent. 95:673-676. Chapman, J.A. and J.M. Kinghorn. 1958. Studies of flight and attack activity of the ambrosia beetle, Trypodendron lineatum (Oliv.) and other scolytids. Can. Ent. 90:362-372. Chapman, J.A. and E.D.A. Dyer. 1960. Characteristics of Douglas-fir logs in relation to ambrosia beetle attack. For. Sci. 15:97-101. Chapman, J.A. and W.W. Nijholt. 1980. Times of log attack by the ambrosia beetle Trypodendron lineatum ambrosia beetle Trypodendron lineatum in coastal British Columbia. Can. For. Serv., Pac. For. Res. Can. Rep. BC-X-207. Daterman, G.E., J.A. Rudinsky, and W.P. Nagel. 1965. Flight patterns of bark and timber beetles associated with coniferous forest of western Oregon. Agricultural Experiment Station. Oregon State University, Corvallis, Oregon. Technical Bulletin 87. Deyrup, Mark. 1981. Deadwood decomposers. Natural History 90(3):84-91. Doane, R.W. and O.J. G i l l i l a n d . 1929. Three California ambrosia beetles. J. Econ. Ent. 22:915-921. 101 Doane, R.W., E.C. Van Dyke, W.J. Chamberlin, and H.E. Burke. 1936. Forest insects. 1st ed., McGraw-Hill Co., New York. Dyer, E.D.A., and J.A. Chapman. 1962. Brood productivity of ambrosia beetle in water soaked logs. Can. For. Serv. Bi-mon. Res. Notes. 2i(3):3. Dyer, E.D.A. 1963a. Distribution of Trypodendron attacks around the circumference of logs. Can. Dep. A g r i c , For. Biol. Div., Bi-monthly. Prog. Rep. 19(2):3-4. Dyer, E.D.A. 1963b. Attack and brood production of ambrosia beetles in logging debris. Can. Entomol. 95:624-631. Dyer, E.D.A. and J.A. Chapman. 1965. Flight and attack of the ambrosia beetle Trypodendron lineatum (Oliv.) in relation to fe l l i n g date of logs. Can. Ent. 97:42-57. Farris, S.H. 1963. Ambrosia fungus storage in two species of Gnathotrichus Eichhoff (Coleoptera:Scolytidae). Can. Ent. 95:257-259. Ford-Robertson, F.C. (ed.). 1971. Terminology of forest science, technology practice and products. Society of American Foresters. 349 pp. Fox, D.J. and K.E. Guirre. 1976. Documentation for Midas. 3rd Ed., Stati s t i c a l Res. Lab., The University of Michigan. 203 pp. Funk, A. 1970. Fungal symbionts of the ambrosia beetle Gnathotrichus  sulcatus. Can. J. Bot. 48:1445-1448. Furniss, R.L. and V.M. Carolin. 1977. Western forest insects. For. Serv. USDA. 654 pp. 1 0 2 Gagne, J.A. and W.H. Kearby. 1979. Life history, development, and insect-host relationships of Xyleborus celsus (Coleoptera: Scolytidae) i n Missouri. Can. Ent. I l l :295-304. Gibson, C. J.M. Kinghorn and J.A. Chapman. 1958. Ambrosia beetle productivity. Can. Dep. A g r i c , For. Biol. Div. Bi-mon. Prog. Rep. 14(5) :2-3). Graham, K. 1959. Release by flight exercise of the chemotropic response from photopositive domination in a scolytid beetle. Nature 184:283-284. Graham, K. 1968. Anaerobic induction of primary chemical attractants for ambrosia beetles. Can. J. Zool. 46:905-908. Graham, K. and E.C. Boyes. 1950. Pinworms in lumber-historical and economical series. B.C. Lumberman 34(8):42, 106. Graham, K. and A.E. Werner. 1956. Chemical aspects of log selection by ambrosia beetles. Can. Dep. A g r i c , For. Biol. Div. Bi-Mon. Prog. Rep. 12(1):3-4. Hadorn, C. 1933. Recherches pur la morphologie, les stades evolutifs et l'hivernage du Bostryche lisere (Xyloterus lineatus Oliv.) (studies on the morphology, developmental stages and hibernation of Bostriches lineatus (Xyloterus lineatus Oliv.). Suppl. aux org. de la soc. forest. Suisse, Bern. 11:1-120. Hosking, G.P. 1972. Xyleborus saxeseni, its life-history and flight behaviour in New Zealand. N.Z. J. of For. Sci. 3_(l):37-53. 103 Jessome, A.P. 1977. Strength and related properties of woods grown in Canada. Eastern Forest Products Laboratory, Ottawa, Ontario. For. Tech. Report 21. 37 pp. Johnson, N.E. 1964. Effects of different drying rates and two insecticides on beetle attacks in felled Douglas-fir and western hemlock. Weyerhaeuser For. Res. Note 58. Johnson, N.E. and J.G. Zing. 1969. Transpirational drying of Douglas-fir: effect on log moisture content and insect attack. J. For. 67(11):816-819. Kinghorn, J.M. 1956. Sapwood moisture in relation to Trypodendron attacks. Can. Dep. A g r i c , For. Biol. Div. Bi-mon. Prog. Rep. JL2(5):3-4. Klinka, K. 1976. Ecosystem units, their classification, interpreta-tion and mapping in the University of British Columbia Research Forest. Ph.D. Thesis, University of British Columbia. 341 pp. Mathers, W.G. 1935. Time of f e l l i n g in relation to injury from ambrosia beetles, or pinworms. B.C. Lumberman. 19:11. McBride, C.F. 1950. The effect of ambrosia beetle damage upon lumber value. B.C. Lumberman 3_4_(9):46-48. McBride, C.F. and J.M. Kinghorn. 1960. Lumber degrade caused by ambrosia beetles. B.C. Lumberman 44(7):40-52. McLean, J.A. 1976. Primary and secondary attraction in Gnathotrichus sulcatus (LeConte) (Coleoptera:Scolytidae) and their application in pest management. Ph.D. Thesis, Simon Fraser University. 108 pp. 104 McLean, J.A. 1980a. Survey using pheromones traps, of Gnathotrichus  sulcatus (Coleoptera:Scolytidae) in two Vancouver Island dryland sorting areas. J. Entomol. Soc. Brit. Columbia 77:20-24. McLean, J.A. 1980b. Tracing the origins of a sawmill population of an ambrosia beetle, Gnathotrichus sulcatus, with X-ray energy spectrometry. Proc. I.U.F.R.O. Conf. Dispersal of Forest Insects: Evaluation, theory and management implications. Idaho 1979. (In press). McLean, J.A. and J.H. Borden. 1975a. Gnathotrichus sulcatus attack. i and breeding in freshly sawn lumber. J. Econ. Entomol. 68:605-6. McLean, J.A. and J.H. Borden. 1975b. Survey of a sawmill populations of Gnathotrichus sulcatus (Coleoptera:Scolytidae) using the pheromone, sulcatol. Can. J. For. Res. 5_: 586-591. McLean, J.A. and J.H. Borden. 1977a. Attack by Gnathotrichus sulcatus (Coleoptera:Scolytidae) on stumps and felled trees baited with sulcatol and ethanol. Can. Ent. 109:675-686. McLean, J.A. and J.H. Borden. 1977b. Suppression of Gnathotrichus sulcatus with sulcatol-baited traps in a commercial sawmill and notes on the occurrence of (Z. retusus and Trypodendron lineatum. Can. J. For. Res. 7^:348-356. McLean, J.A. and J.H. Borden. 1979. An operational pheromone-based suppression program for an ambrosia beetle, Gnathotrichus  sulcatus, in a commercial sawmill. J. Econ. Entomol. 72:165-72. 105 Milligan, R.H. 1969. An introduced ambrosia beetle (Xyleborus saxeseni) attacking logs and freshly sawn lumber. N.Z. Serv. Res. Leaflet 22. 4 pp. Milligan, R.H. 1970. Overseas wood- and bark-boring insects inter-cepted at New Zealand ports. Tech. Pap. N.Z. For. Serv. 57. 80 pp. Mizzel, Russel F. and T. Evan Nebecker. 1978. Estimating the developmental time of the southern pine beetle Dendroctonus  frontalis as a function of f i e l d temperatures. Environ. Entomol. 7_:592-595. Moeck, H.A. 1971. Field test of ethanol as a scolytid attractant. Dep. Fish. For. Bi-Mon. Res. Notes 27(2):11-12. Moeck, H.A. 1980. Host selection behaviour of bark beetles (Coleoptera:Scolytidae) attacking Pinus ponderosa, with special emphasis on the western pine beetle. Ph.D. Thesis, University of California, Berkeley. Nijholt, W.W. 1973. The effect of male Trypodendron lineatum (Coleoptera:Scolytidae) on the response of f i e l d populations to secondary attraction. Can. Entomol. 105:583-590. Nijholt, W.W. 1978. Ambrosia beetle. A menace to the forest industry. Can. For. Serv., Pac. For. Res. Cen. Rep. BC-P-25. Nijholt, W.W. 1979. The striped ambrosia beetle Trypodendron lineatum (Olivier). An Annotated Bibliography. Can. For. Serv., Pac. For. Res. Cen. Rep. BC-X-121. 106 Panshin, A.J. and Carl de Zeeuw. 1980. Textbook of wood technology. McGraw-Hill Book Company. 722 pp. Parker, M.L., W.G. Warren, and R.W. Kennedy. 1976. Effect of thinning and f e r t i l i z a t i o n on intra-ring characteristics and kraft pulp yield of Douglas-fir. Applied Polymer Symposium No. 28, 1075-1086. John Wiley & Sons, Inc. Person, Hubert L. 1931. Theory in explanation of the selection of certain trees by the western pine beetle. J. For. 29:696-699. Prebble, M.L. and K. Graham. 1957. Studies of attack by ambrosia beetles in softwood logs on Vancouver Island. For. Sci. H3):90-112. Rudinsky, J.A. 1966. Scolytid beetles associated with Douglas-fir. Response to terpenes. Science 152:218-219. Rudinsky, J.A. and I.A. Schneider. 1969. Effects of light intensity on the fl i g h t pattern of two Gnathotrichus (Coleoptera: Scolytidae) species. Can. Ent. 101:1248-1255. Skoosgaard, J. 1968. Douglas-fir beetle in British Columbia. Can. For. Serv., FIDS. Leaflet 14. 5 pp. Walters, J., J. Soos and P.G. Haddock. 1960. The selection of plus trees on the University of British Columbia Forest, Haney, B.C. Faculty of Forestry, UBC. 12 pp. Wilson, J.W. 1968. Organization within the coniferous growth ring. Pulp and Paper Res. Inst. Can. Woodlands Rep. W.R. No. 7. Winer, B.J. 1971. S t a t i s t i c a l principles in experimental designs. 2nd ed. McGraw-Hill Book Company. 907 pp. 107 APPENDIX 1: Results of the analysis of variance of Gnathotrichus sulcatus attacks/m 2 on western hemlock and Douglas-fir stumps in the UBC Research, Forest, 1978-79. Source of 19781 19791 Variation D.F. Mean square F D.F. Mean square F Site 1 • 6.043 16.64**2 1 0.0285 0.06 Host 1 2.181 6.01* 1 2.1478 4.71* Site x host 1 0.02154 0.06 1 0.00000976 0.00 Error 12 0.3631 12 0.4556 ^Data transformed to X' = logio(X + 1) 2Significance levels indicated: * = P <0.05, ** = p <0.01. 108 APPENDIX 2: Results of the analysis of variance of Gnathotrichus sulcatus attacks/m 2 on western hemlock and Douglas-fir trunks in the UBC Research Forest, 1978-79. 19781 19791 Source of Variation D > F > M e a T 1 F D.F. Mean square square Site Host Condition Site x Host Site x Condition Host x Condition Site x Host x Condition Error 2.1767 73.81**' 1.0413 35.31** 0.1820 6.17* 0.09772 3.31 0.3008 10.20* 0.1125 3.81 0.000391 0.013 8 0.02949 0.1989 0.62 0.01663 0.05 0.3545 1.11 0.0349 0.11 0.5109 1.60 0.4789 1.50 0.03118 0.10 8 0.3190 *Data transformed to X' = logio(X + 1) S i g n i f i c a n c e l e v e l s indicated: * = P <0.05; ** = P <0. 109 APPENDIX 3: Results of the analysis of variance for percent of Gnathotrichus sulcatus attacks per quadrant on 1 m long western hemlock and Douglas-fir logs. UBC Research Forest, Maple Ridge, B.C. May-November, 1978 and March-October, 1979. N = 8 per each species. WESTERN HEMLOCK DOUGLAS-FIR Source of Variation, D > F > M e a n F Mean square square Percent of attacks per quadrant* 3 589.72 6.23**2 310.28 0.30 Aspect Error 28 94.65 244.55 iRaw data. No transformation as Bartlett's Test showed homogeneous variances. P r o b a b i l i t y l e v e l indicated: ** = P <0.01. f Yv,rk beetles total attacks on Source of Variation D.F. Site Condition Site x Condition Error 1 4 WESTERN HEMLOCK P. tsugae Mean square 1.141 0.014 0.0007 0.248 4.59 4.59 0.003 S. tsugae Mean square 0.079 0.758 0.772 0.130 0.61 5.84 5.94 l D a t a transformed to X' - l o g 1 0 ( X + U 2 P r o b a b i l i t y level indicated: * - * <0-05' DOUGLAS-FIR D. pseudotsugae Mean square 0.734 0.367 0.003 0.022 3.37 1.69 0.01 S. nnlspinosus Mean F square 0.049 3.652 0.189 0.436 0.11 8.37*2 0.43 Ill APPENDIX 5: Results of the analysis of variance of Trypodendron lineatum attacks/ro.2 o n w e s t e r n hemlock and Douglas-f i r trunks in the UBC Research Forest, 1978-79. Source of Variation D.F. 1 Mean square F Site 0.536 1.51 Host 1 0.352 0.99 Site x Host 1 0.0523 0.15 Condition 1 0.107 0.30 Site x Condition 1 0.481 1.36 Host x Condition 1 0.0163 1.78 Error 8 0.354 1Data transformed to X' = logifj( x + 112 APPENDIX 6: Results of the analysis of variance of brood (BR) APPbNUlA o. *es ( f r ) o f G ^ a t h o t r l c h u s sulcatus galleries on western hemlock and Douglas-fir initiated during 1978-79, UBC Research Forest, Maple Ridge, B.C. BR1 FR1 Source of Variation M e a n Mean D , F * square square TREES: Site 1 0.01966 0.13 0.2538 1.76 Host 1 7.6047 50.58** 11.968 83.23** Site x Host 1 0.1473 0.98 0.0953 0.66 Condition 1 0.0261 0.17 1.9857 13.81** Site x Condition 1 0.1906 1.27 0.000027 0.00019 Host x Condition 1 1.764 11.73** 3.5204 24.48** Site x Host x Cond. 1 1.794 11.93** 0.807 5.62* Tree (Site,Host,Cond) 8 0.3360 2.23* 0.5174 3.60** Error 61 0.1503 0.1438 STUMP: Site 1 0.3040 3.01 0.6475 4.93* Host 1 9.4935 94.29** 6.7968 51.8** Site x Host 1 0.3570 3.55 0.3400 2.59 Stump (Site,Host) 10 0.4787 4.75** 0.2813 2.14* Error 56 0.1007 0.1312 *Data transformed to X' - l o g 1 0 ( X + ^ S i g n i f i c a n c e l e v e l s indicated: * = P <0.05; APPENDIX 7: Results of the analysis of variance for Gnathotrichus sulcatus productivity/ APPEND IX l l e r y a n d m e a n gallery parameters in western hemloc^ana bouglas-fir logs in laboratory conditions. Faculty of Forestry, UBC 1978-79. Source of Variance EN Host Tree D.F. Mean square PN Mean square BR Mean square LG Mean square DP Mean square 1 28.9 3.58*2 592.9 11.9** 756.9 13.32** 97.34 0.76 0.016 0.007 2 0.80 0.099 13.07 0.25 8.60 0.15 136.8 1.07 5.70 2.56 Error 16 8.07 52.97 56.82 127.9 2.22 *Data transformed to X' = log^CX + ^ S i g n i f i c a n c e levels indicated: * = P <0.05; ** - P <0.01. 114 APPENDIX 8: Results of the analysis of variance of mean numbers of egg niches (EN) and pupal niches (PN) of Gnathotrichus sulcatus galleries on western hemlock and Douglas-fir initiated during 1978-79, UBC Research Forest, Maple Ridge, B.C. EN1 PN1 Source of Variation Mean Mean D , F * square F square TREES: Site 1 0.0068 0.08 0.07650 0.048 Host 1 0.285 3.20 9.1018 57.61** Site x Host 1 0.0003 0.004 0.3042 1.93 Condition 1 0.133 1.50 0.00701 0.44 Site x Condition 1 0.010 0.12 0.1077 0.68 Host x Condition 1 0.0054 0.06 1.5181 9.61** Site x Host x Cond. 1 0.075 0.84 2.2443 14.21** Tree (Site,Host,Cond) 8 0.233 2.62 0.4702 2.98** Error 61 0.089 0.1580 STUMP: Site 1 0.000194 0.054 0.1200 0.98 Host 1 0.05864 1.65 7.3561 59.60* Site x Host 1 0.03613 1.01 0.2750 2.06 Stump (Site,Host) 10 0.3139 8.81** 0.6768 2.09** Error 56 0.3563 0.1333 1Data transformed to X' = log 1Q(X + 1) 2 Significance levels indicated: * = P <0.05; ** « P <0.01. 115 APPENDIX 9: Results of the analysis of variance of mean length (LG) and depth of penetration (DP) of Gnathotrichus sulcatus galleries on western hemlock and Douglas-fir initiated during 1978-79, UBC Research Forest, Maple Ridge, B.C. LG 1 DP1 Source of Variation Mean Mean D , F ' square F square TREES: Site 1 0.003436 0.053 0.1019 8.53** Host 1 6.8474 105.79**2 2.5002 210.79** Site x Host 1 0.1479 2.28 0.02613 2.20 Condition 1 0.7155 11.05** 0.09059 7.63** Site x Condition 1 0.01061 0.164 0.02304 1.94 Host x Condition 1 1.2083 18.67** 0.03910 3.30 Site x Host x Condi 1 0.1420 1.19 0.006688 0.56 Tree (Site,Host,Cond) 8 0.2251 3.48** 0.04329 3.65** Error 61 0.064728 0.001186 STUMP: Site 1 0.02081 0.21 0.00917 0.26 Host 1 3.2892 33.74** 1.6745 48.32** Site x Host 1 0.1404 1.44 0.09328 2.69 Stump (Site,Host) 10 0.2702 2.77 0.04931 1.42 Error 56 0.09750 0.03465 *Data transformed to X' » l o g 1 0 ( X + D S i g n i f i c a n c e l e v e l s indicated: * = P <0.05; ** = P <0.01. APPENDIX 10: Results of the analysis of variance of mean number of egg niches (EN), pupal niches (PN), depth of penetration (DP) and length (LG) of Gnathotrichus sulcatus galleries in western hemlock trunks over 13 months. UBC Research Forest, Maple Ridge, B.C. 1978-79. Source of Variance EN1 PN1 DP1 LG 1 D.F. Mean square F Mean square F Mean square F Mean square F Site 1 0.0096 0.16 0.600 2.85 0.154 7.20** 0.416 4.14* Cond. 1 0.0086 0.14 2.129 1 0 . i i * * 2 0.025 1.19 0.013 0.13 Month 12 0.068 1.13 0.140 0.67 0.034 1.61 0.100 1.00 Site x Cond. 1 0.115 1.91 0.735 3.43 0.0088 0.41 0.013 0.13 Site x Month 9 0.072 1.19 0.455 2.16* 0.030 1.39 0.176 1.76 Cond. x Month 10 0.080 1.32 0.641 3.04** 0.022 1.02 0.389 3.88* Site x Cond. x Month 6 0.086 1.42 0.171 0.81 0.032 1.50 0.200 1.99 Tree (Site,Cond., Month) 34 0.115 1.91** 0.400 1.89** 0.039 1.81* 0.210 2.09** Error 66 0.060 0.211 0.021 0.100 ^Data transformed to X' = l o g i Q ( X + 1) S i g n i f i c a n c e levels indicated: * = P'<0.05; ** = P <0.01. ™ x^^^^^s^^^^^^t (PN)' galleries in Douglas-fir trunks over 13 months. UBC Research Forest, Maple Ridge, B.C. 1978-79. Source of Variance D.F. EN Mean square PN Mean square DP Mean square LG Mean square Site Cond. Month Site x Cond. Site x Month Cond. x Month Site x Cond. x Month Tree (Site,Cond, Month) Error 1 0.047 0.74 1.064 5.20* 0.021 1.44 0.335 1.96 1 0.330 5.16*2 0.621 3.03 0.376 2.52 0.030 0.17 12 0.051 0.81 0.449 2.19* 0.015 1.03 0.245 1.44 1 0.183 2.86 0.671 3.27 0.0378 2.46 1.05 6.14** 11 0.138 2.17* 0.324 1.58 0.013 0.84 0.159 0.93 12 0.102 1.60 0.285 1.39 0.024 1.60 0.087 0.51 7 0.106 1.66 0.213 1.04 0.024 1.60 0.086 0.51 28 0.091 1.43 0.340 1.66 0.032 2.12** 0.189 1.11 54 0.063 0.205 0.015 0.171 iData transformed to X' = l o g 1 Q ( X + 1) S i g n i f i c a n c e l e v e l s indicated: * = P <0.05; ** = P <0.01. 118 APPENDIX 12: Results of the analysis of variance for Gnathotrichus sulcatus catches on bark phloem, sapwood and heartwood stimuli prepared from western hemlock and Douglas-fir. Set out in Point Roberts, Washington on June and July, 1978, Source of Variation Degrees of Mean freedom square Western hemlock experiment Week Sex Block Treatment Week x block Block x treatment 3 Error 2 1 4 3 8 12 89 0.094 0.192 0.101 0.218 0.044 0.088 0.018 5.07**' 10.35** 5.45* 11.74** 2.39* 4.76* 1 Douglas-fir experiment Sex Block Treatment 3 Error 1 2 4 22 0.0715 0.0026 0.536 0.0142 5.00* 0.18 37.49** *Data transformed to X' = log 1 Q(X + ^ Significance levels indicated: * = P <0.05; ** = P <0.01. ^ A l l other interactions were pooled in the error term because they were non-significant. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0095129/manifest

Comment

Related Items