@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Land and Food Systems, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Watson, Alan Kemball"@en ; dcterms:issued "2011-04-15T23:59:04Z"@en, "1972"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Centaurea diffusa was found to infest 64,079 acres in the semiarid interior of southern British Columbia. C. maculosa has infested an additional 8,420 acres. These weedy plant species are common along roadsides-and in waste places and are spreading rapidly over vast areas of semiarid rangeland. The density of the knapweed species could only be correlated with the degree of soil disturbance and was not correlated with any chemical property of the soil. The knapweeds are not generally grazed by domestic livestock and substantially reduce the forage yields in heavily infested areas. C. diffusa and C. maculosa prevent soil erosion on disturbed sites as these pioneer species are capable of rapid colonization of these sites. These species are utilized to some extent as honey plants. However, the losses attributed to knapweed infestations generally override any potential use of the knapweeds. Seeds of C. diffusa and C. maculosa germinate readily over a wide range of conditions. Continuous light was found to significantly retard the gemination of both species. The optimum temperature for germination of C. maculosa was found to be lower than that for C. diffusa seed. The optimum soil depth for emergence of both species was on the soil surface. C. diffusa seed did not emerge from soil depths below 3 cm. C. maculosa seed emerged from a soil depth of 5 cm. Phenological data were recorded for both species. C. maculosa begins to flower in early July with C. diffusa flowering approximately one or two weeks later. The annual reproductive capacity of C. diffusa was determined to be 665 with C. maculosa exhibiting an annual reproductive capacity of 298. Dual mechanisms of seed dispersal were observed for both species. Propagule dispersal is mainly by wind. Cultural methods of control were generally found not to be beneficial in controlling the knapweeds. The knapweeds can be adequately controlled by using the herbicide picloram, 4-.amino - 3,5,6 - trichloropicolinic acid, however, the high cost of this herbicide has greatly limited its use. Phytophagous insects, Urophora affinis and Metzneria paucipunctella, have been studied as potential biocontrol agents for the knapweeds. Initial releases of Urophora affinis indicate that the insect is capable of survival in the southern interior of British Columbia. The population has increased substantially since the initial release in 1970. It will take up to 10 years or more, however, before the releases of U. affinis are sufficiently well established to reduce the knapweed infestations. Two fungal organisms, Sclerotinia sclerotiorium and Microsphaeropsis centaureae, were isolated from diseased C. diffusa in the Vernon area. The potential of these organisms as biotic agents has, as yet, not been determined. Weed control methods must be associated with other appropriate management practices to produce increases in forage yields. Such an integrated approach to the control of diffuse and spotted knapweed will substantially reduce the extent of the knapweed infestation in the southern interior of. British Columbia."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/33706?expand=metadata"@en ; skos:note "THE BIOLOGY AND CONTROL OF CENTAUREA DIFFUSA LAM. AND CENTAUREA MACULOSA LAM. by-Alan Kemball Watson B.Sc. (Agr.) University of British Columbia 1970 A Thesis Submitted i n Partial Fulfilment of the Requirements for the Degree of Master of Science in the Department of Plant Science / accept this thesis as conforming to the required standard The University of British Columbia May, 1972 In present ing th i s thes is in. pa r t i a l f u l f i lmen t o f the requirements fo r an advanced degree at the Un ive rs i t y of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f r ee l y ava i l ab le for reference and study. I fu r ther agree that permission for extensive copying of th i s thes i s for s cho la r l y purposes may be granted by the Head of my Department or by his representat ives . It is understood that copying or pub l i c a t i on of th is thes is fo r f i nanc ia l gain sha l l not be allowed without my Written permiss ion. Department of The Un ivers i ty of B r i t i s h Columbia Vancouver 8, Canada Date 2-^ (i) ABSTRACT Centaurea diffusa was found to infest 64,079 acres in the semiarid interior of southern Brit i s h Columbia. C. maculosa has infested an additional 8,420 acres. These weedy plant species are common along roadsides-and i n waste places and are spreading rapidly over vast areas of semiarid rangeland. The density of the knapweed species could only be correlated with the degree of s o i l disturbance and was not correlated with any chemical property of the s o i l . The knapweeds are not generally grazed by domestic livestock and substantially reduce the forage yields i n heavily infested areas. C. diffusa and C. maculosa prevent s o i l erosion on disturbed sites as these pioneer species are capable of rapid colonization of these sites. These species are u t i l i z e d to some extent as honey plants. However, the losses attributed to knapweed infestations generally override any potential use of the knapweeds. Seeds of C. diffusa and C. maculosa germinate readily over a wide range of conditions. Continuous light was found to significantly retard the gemination of both species. The optimum temperature for germination of C. maculosa was found to be lower than that for C. diffusa seed. The optimum s o i l depth for emergence of both species was on the s o i l surface. C. diffusa seed did not emerge from s o i l depths below 3 cm. C. maculosa seed emerged from a s o i l depth of 5 cm. Phenological data were recorded for both species. C_. maculosa begins to flower i n early July with C. diffusa flowering approximately one or two weeks later. The annual reproductive capacity of C. diffusa was determined to be 665 with C. maculosa exhibiting an annual reproductive capacity of 298. ( i i ) Dual mechanisms of seed dispersal were observed for both species. Propagule dispersal is mainly by wind. Cultural methods of control were generally found not to be beneficial in controlling the knapweeds. The knapweeds can be adequately controlled by using the herbicide picloram, 4-.amino - 3,5,6 - trichloropicolinic acid, however, the high cost of this herbicide has greatly limited i t s use. Phytophagous.insects, Urophora a f f i n i s and Metzneria paucipunctella, have been studied as potential biocontrol agents for the knapweeds. I n i t i a l releases of Urophora a f f i n i s indicate that the insect i s capable of survival in the southern interior of British Columbia. The population has increased substantially since the i n i t i a l release i n 1970. It w i l l take up to 10 years or more, however, before the releases of U. affinis are sufficiently well established to reduce the knapweed infestations. Two fungal organisms, Sclerotinia sclerotiorium and Microsphaeropsis centaureae, were isolated from diseased C_. diffusa i n the Vernon area. The potential of these organisms as biotic agents has, as yet, not been determined. Weed control methods must be associated with other appropriate manage-ment practices to produce increases in forage yields. Such an integrated approach to the control of diffuse and spotted knapweed w i l l substantially reduce the extent of the knapweed infestation i n the southern interior of. British Columbia. ( i i i ) TABLE OF CONTENTS Page ABSTRACT • i LIST OF TABLES - ........... i v LIST OF FIGURES ........ • v i i ACKNOWLEDGEMENT , ........ •••••• v i i i I. INTRODUCTION • • • 1 II. DISTRIBUTION 2 A. Literature Review ....... .• 2 B. Experimental Methods 5 C. Results and Discussion ...... 7 I I I . WEED ECOLOGY ... : 16 A. Literature Review •.... 16 B. Experimental Methods • • • • • 28 C. Results and Discussion • • . 33 IV. CONTROL , 51 A. Literature Review 51 B. Experimental Methods ••• • , 70 C. Results and Discussion ' 72 V. BIBLIOGRAPHY 91 VI. APPENDIX • • 100 (iv) LIST. OF TABLES,. Table Page I Density scale u t i l i z e d i n the Centaurea survey ' 5 II Methods of s o i l analysis ,• 6 III. Acreage infested with Centaurea species i n B.C 7 IV Plant density of knapweed infestations 9 V Average monthly and annual mean temperatures of selected sites .. .• ,.• 10 . VI Average monthly and annual precipitation and altitude of selected sites ', 11 VII Precipitation and mean temperatures of selected sites i n 1971 ... . 12 VIII Chemical analysis of knapweed infested soils 14 IX Chemical composition of the above ground parts of Centaurea diffusa .(% dry wt.) ............. 21 X Chemical analysis of C. diffusa and C. maculosa i n flower (% dry wt. ) .... . •. / 22 • XI Effect of C_. maculosa water extract on crop seedlings (7 days after sowing) 27 XII Effect of C. diffusa water extract on crop seedlings (7 days after sowing) ... ... ... 28 XIII Yield of forage and knapweed from knapweed infested rangeland i n the southern interior of British Columbia 34 XIV Weight of C. diffusa and C_. maculosa seed gathered i n British Columbia' , 39 XV The effect of light on germination of C. diffusa -and C. maculosa seed (adjusted means).. 40 XVI Optimum germination temperatures for C. diffusa-and C. maculosa ,. • 43 XVII • Average seed production of C. ..diffusa and C. maculosa . .•. 46 XVIII Shallow plowing (7 cm) on C. diffusa i n the fruiting stage (% cover) .... 51 , ( v ) Table Page XIX- Plowing- (18 cm) on C. diffusa in the fruiting stage (%' cover) ._. -. 52 XX • Mowing of C. diffusa in the bolted stage (% cover) ..... 52 XXI Effect of burning on C. diffusa in the fruiting stage (% cover) .. .. ..' 53 / • . . . XXII Effect of picloram, reseeding, and complete protection from grazing by domestic animals on the control of C. - diffusa ..< ... • : • 54 XXIII 2,4-D (1 kg/hectare) applied to pasture infested with with diffuse knapweed in the rosette stage (% cover) . 54 XXIV Potential insect agents for the biological control of C. diffusa , ..... 59 XXV, Potential insect.agents for the biological control of C. maculosa ........ 60 XXVI Factors and responses involved in the host specificity of ovippsition in U. affinis • • •' • 63 XXVII Length of ovipositor; .and dimensions of oviposition site of some Urophora spp. • •••• ........ 64 XXVIII Urophora affinis. attack of Centaurea at B.C. release sites' ........... • 66 XXIX Effects of soil disturbance on C. diffusa populations ... 72 XXX Shallow plowing with later fertilizer application (10T fresh horse manure/hectare) (% cover) 73 XXXI The effects of mowing on knapweed • 74 XXXII Results of the East Kootenay Knapweed Control Program oc 1970-1971 75 XXXIII Treatments utilized to break sclerotia dormancy ? 80 XXXIV Determination of pathogenicity of Sclerotinia sclerotiorium and Microsphaerosis centaureae- on Centaurea diffusa and C. maculosa .................... 82 XXXV- Plant diseases observed on Centaurea species in North America 777777777. ,83 (vi) Table - . Page XXXVI Evaluation of biological control measures against weeds i n Canada ' 87 Appendix Table I. Effect of temperature on germination of C. diffusa and C. maculosa ...................... 100 Appendix Table II . . Effect of sowing depth on seedling emergence of C. diffusa and C. maculosa • 101 ( v i i ) LIST OF FIGURES Figure Page 1 Distribution of C. diffusa i n the U.S.A 4 2 Distribution of C.\\ maculosa i n the U.S.A. . .• 4 3 Distribution of C. diffusa i n British Columbia .. 8 4 Distribution of C. maculosa in 'British Columbia ;. 8 5... Temperature gradient bar ,. 31 6j C. diffusa and C. maculosa seeds on temperature gradient bar 31 7 Pollen grains of C. nigra, C. maculosa and C. diffusa ... 38 8 C. maculosa, (upper) and C. diffusa seeds .. .. 38 9 The effect.of temperature on germination of C. diffusa and C. maculosa seed • 42 10 Seedling emergence of C. diffusa and C. maculosa 24 days after sowing at different s o i l depths ........ 43 11 C. maculosa, (left) and C. diffusa flower heads ......... 50 12 Fusiform gall i n the receptacle of a C. maculosa flower, head , 61 13 Sclerotium i n a C. diffusa, root 61 14 Apothecia produced from cold treated sclerotia 79 15 Plate culture of S_. sclerotiorium. 79 16 Plate culture of M. centaureae 81 17 Leaf spot on C. diffusa 85 18 Leaf spot on C. maculosa 85 19 Re-establishment of C. diffusa-after herbicide treat-ment at OK Fa l l s , B.C 89 . 20 Re-establishment'of C. maculosa after herbicide treat-ment 'at Chase, B.C. •••• • 89 ( v i i i ) . ACKNOWLEDGEMENT I thank Dr. A.J. Renney for his interest and supervision throughout this study. The assistance of the other members of the committee: Dr. V.C. Runeckles, Dr. W.G. Wellington and Dr. R.J.'.Copeman i s gratefully appreciated. •In particular,.I am indepted to Dr..'R.J. Copeman for advice and materials in regard to the work with the fungal organisms. • I also thank Dr. G.W. Eaton, for assistance with s t a t i s t i c a l problems and Dr. R. Taylor for obtaining herbarium specimens. I thank Dr..R.J. Bandoni, Department of Botany, University of British Columbia; Miss M.E. E l l i o t t , Plant Research Institute, Ottawa; and Dr. G. Morgan-Jones, Department of Biology, University of Waterloo for their valuable assistance in the identification of the fungal isolates. I gratefully appreciate the co-operation of Dr. J.E. Miltimore, W.A. Hubbard'''and other members of the staff at the Canada Department of Agriculture Research Stations at Kamloops and Summerland. The interest and advice of Dr. P. Harris,.Research Institute, B e l l e v i l l e , Ontario i s gratefully acknowledged. I thank F.'G. Smith, Department of Transport, Gonzales Observatory, Victoria for making weather data readily available. The assistance of the following British Columbia Department of Agriculture staff i s acknowledged: E.L. Berry, J.C. Ryder, J. Corner and A.H. Bawtree. The technical assistance of Miss M. Johnston, Miss A. Harris, I. Derics and J. Gibson is appreciated. The co-operation of numerous ranchers i n the Province i s gratefully acknowledged. I thank Miss P. Jenkinson for the typing of the manuscript. The work was supported by the Bostock Grant'and by the National Research Council of Canada., I. INTRODUCTION Centaurea diffusa Lam. (diffuse knapweed) and Centaurea maculosa Lam. (spotted knapweed) are two very troublesome weeds i n the dry southern interior of the province of British Columbia. These aggressive, alien, pioneer species readily colonize disturbed sites and seemingly have the innate a b i l i t y to withstand the competi-tion of the surrounding grass and herbage species and establish almost solid stands of• single species vegetation. Diffuse and spotted knapweed are common along roadsides and on waste ground. They have also- spread onto many acres of overgrazed or otherwise misused rangeland resulting i n marked reductions i n the productive capacity of these ranges. These weeds generally do not yield to the commonly . used weed control practices. The following study was undertaken to determine the extent of the knapweed infestation i n the southern interior of the province of British Columbia, to develop adequate control procedures for C. diffusa and C. maculosa and to develop a more comprehensive understanding of the biology of these two knapweed species. 2 II . . DISTRIBUTION A. Literature Review The earliest western collection of Centaurea diffusa Lam. was made i n an a l f a l f a f i e l d at Bingen, Klickitat County, Washington, i n 1907 by Suksdorf (39,59). The f i r s t British Columbia record was at Oyama i n 1936 by, Tisdale followed by subsequent collections i n 1939-40 by Eastham at Penticton and Grand Forks (9,12,33,93,9,5). However, Renney (71) pointed out that diffuse knapweed infestations occurred prior to 1930 as the weed was observed at Lytton and Pritchard at this time. Eastham (19) described the distribution of C. diffusa i n .1947 as abundant i n the Okanagan and spreading i n the Grand Forks d i s t r i c t . Renney, i n 1959, described C. diffusa as being \"well adapted to the dry interior\" and indicated that the weed had infested the Fraser Canyon around Lytton, the Okanagan Valley, the Grand Forks area and the Cranbrook-Kimberly area. Since that time C_. diffusa has spread rapidly and colonized large acreages of dry land ranges and roadsides i n the interior of British Columbia. The f i r s t Canadian collection of Centaurea maculosa Lam. was made i n 1893 at Victoria by Maeoun as C. paniculata (33). C. maculosa i s very abundant along roadsides and waste places i n the Canford area, near Merrit, on the west arm of Kootenay Lake and i n the Slocan area (19,71). As i s the case with C. diffusa., C. maculosa has also spread rapidly. Frankton (28) has indicated that the Canadian distribution of C. diffusa is. limited to southern British Columbia. Meanwhile, 3 C. maculosa has a wider Canadian distribution with large infestations i n Ontario and Nova Scotia, but i s most abundant in British Columbia (28). C. maculosa and C. diffusa have a broad distribution in the United States of America and significant, increases in knapweed infesta-tions have been observed in the'.past ten years. C. diffusa has been observed commonly i n f i e l d s , along .roadsides and in waste places in Washington, Oregon, Idaho, I l l i n o i s , Missouri, Iowa, Michigan, and Massachusetts (1,10,16,24,31,48,49,62,66,74,75,101). C. maculosa has a wider distribution than C_. diffusa i n the U.S. and has been recorded in Washington, Oregon, Idaho, Montana, Massachusetts, New Jersey', Pennsylvanhia, I l l i n o i s , Kansas, Minnesota, Missouri, South Dakota,. -Michigan, Indiana, Vermont, Tennessee and North Carolina (1,10,16,17, 24,31,48,49,50,62,66,74,75,87,88 ,89',101). . C. maculosa i s commonly found along roadsides, i n waste places, i n fields and in rangeland i n the United States of America. Figures 1 and 2 ill u s t r a t e the distribution °f £• diffusa and C.. maculosa i n the U.S.A. Roche (74) has indicated that a total of 378,585 acres i n the State of Washington are infested with C_. diffusa with control measures having been attempted on only 6% of the area infested. Popova (67) indicates that C. diffusa has a substantial area of distribution in eastern Europe and Asia. C. diffusa i s common in Romania, Yugoslavia, northern Italy, eastern shore of the Mediterranean, Turkey, Greece, Bulgaria, Asia Minor and Syria. C_. diffusa is common in the U.S.S.R., especially the\" Ukraine and the Crimea (67). 4 Fig.. 2. Distribution of C, maculosa in the 'J,S.A, (U.S.D.A. AgriTltodboofc (denser hatching - weed of greater economic importance) 5 B. . Experimental Methods A recent survey of the southern interior of British Columbia has resulted i n a comprehensive understanding of the distribution of these two weedy Centaurea species. The knapweed survey included the following descriptions of each infestation; species, density, area, acreage, and location. A density scale, based on the Bran-Blanquet (8) system, was established to characterize the weed infestations according to plant frequency (Table I ) . The particular \"area\" where the infestation occurred was defined as follows: roadside verge - V; railroad rights-of-way - RR; rangeland - R; and f i e l d - F. TABLE I.. 'Density Scale Utilized i n the Centaurea Survey C. diffusa x '• \" 1 plant per 1/8 acre 1 1 plant per 1/8 acre to 1 plant per 100- f t 2 patches 3 1 to 110 plants per 100 f t 2 4 10 to 100 plants per yd 2 . . . 2 ... 5 more than 100 plants per yd C. maculosa x 1 plant per 1/8 acre 1 1 plant per 1/8 acre to 1 plant per 100 f t 2 patches 3 1 to 11 plants per 100 f t 2 2 4 1 to 30 plants per yd 2 5 more than 30 plants per yd 6 With the distribution of C. diffusa and C. maculosa having been established within the province of British Columbia, analysis of some environmental conditions was attempted to explain the distribution of- . these two weedy plant species. Plant density of the knapweed species varied considerably i n the extensively studied areas i n the Okanagan-Kamloops region. Soil samples from these sites were analyzed i n an attempt to explain the variable plant densities observed i n the C. diffusa and C. maculosa populations. Four s o i l samples were collected from each of eight . diffusa locations and three maculosa locations. The samples consisted of five -random 2 cm. diameter cores, 23 cm. i n length, taken from within random meter square quadrats. The techniques u t i l i z e d for the s o i l analysis are l i s t e d i n Table' I I . Data were analyzed by analysis of variance and simple linear multiple regression. TABLE I I . Methods of Soil Analysis Measurement Method pH 1:5 soil:water suspension (pH meter) pH 0.01 M CaCl„ solution (pH meter) conductivity saturated paste - (conductivity meter) Nitrogen Semi-micro Kjeldahl Available phosphorus Bray P^ technique Sulphur Laboratory Equipment Corporation (LECO) analyzer Carbon Laboratory Equipment Corporation (LECO) analyzer Exchangeable cations Leaching with MLOAc (pH 7) and KC1 and u t i l i -(Na, K, Mg, Ca) zation of the Perkin-Elmer Flame Spectrophotometer Total Exchange D i s t i l l a t i o n with semi-micro Kjeldahl Capacity Texture Hand method based on U.S.D.A. Textural Triangle 7 C. Results and Discussion Figures 3 and 4 il l u s t r a t e the distribution of C_. diffusa and C. maculosa i n British Columbia. An estimate of the acreages infested by Centaurea species i n the province of British Columbia i s given i n Table I I I . TABLE I I I . Acreage infested with Centaurea species in B.C. Species Acres Centaurea diffusa 64,079 C. maculosa 8,420 C. repens 109 C. nigra (C. x pratensis) < \\ The percentage of the infestations of C. diffusa,and C_. maculosa having a particular density classification (Table IV) indicate the capability for spread of both species and the establishment of numerous dense stands of Centaurea species. i Fig. 3. Distribution of C_.' diffusa in British Columbia Fig. 4. Distribution of C. maculosa i n British Columbia 9 TABLE IV. Plant density of knapweed infestations Density C. diffusa C. maculosa X 16.0% 19.0% 1 8.5 19.5 2 3.5 0.5 3 16.0 16.0 4 33.0 19.0 5 23.0 26.0 Popova (67) suggested that the northern boundary of distribution of C. diffusa i s approximately 53°N Lat., however, i t extends further north only along railroad rights-of-way. This report suggests that C. diffusa, has a potential for extending farther north into the Cariboo area of British Columbia since -the northern li m i t of C. diffusa for this province i s presently approximately 51°N. Lat. Both Centaurea species appear to be well adapted to the climatic range of the southern interior of the province (Tables V, VI, .VII). C. diffusa has been observed at altitudes ranging from 500 to over 3,000 feet above sea level. Similarly, C. maculosa has been observed from 100 to over 4,000 feet above sea level. However, both species are most commonly found between 1,000 and 2,000.feet above sea level i n the southern interior of the province. • C. maculosa has a more northern limi t than.C. diffusa with a corresponding lower annual mean temperature range, 43°F-46°F. as compared to 45°F-49°F for the study sites i n the TABLE V. Monthly and Annual Mean Temperatures (Average) Jan. Feb. March Apr i l May June July Aug. Sept. Oct. Nov. Dec. Annual Years Averaged Oliver Penticton (A) .'; Summerland (C.D.A.) Vernon Vernon , (Coldstream) Chase Falkland Kamloops (C.D.A.) Sicamous Westwold 26 27 26 25 22 25 23 '23 25 19 31 31 30 32 26 31 30 31 29 24 40 39 39 36 36 36 34 38 36 34 51 '48 49 47 47 47 46 49 46 44 59 65 57 63 57 63 54 62 56 61 58 65 71 68 70 68 67 71 69 66 67 66 64 55 62 68 66 53 61 • 66 64 68 60 58 60 56 47 57 54 60 49 37 31 48 37 49 36 47 35 48 36 32 30 27 46 34 27 46 35 29 45 33 27 29 56 63 . 69 66 57 46 .34 28 53 59 64 61 54 43 31 25 49 48 48 46 45 46 45 48 46 43 10 7 15 11 \"Thirty year standard period average 1931-60. Source: TDepartment of Transport, Gonzales^Qbservatory,..Victoria,- B. C. TABLE V I . Average monthly and annual precipitation and altitude of selected sites A p r i l -Jan. Feb. March April May June July Aug. Sept. Oct. Nov. Dec. Annual Altitude Total Oliver 1. .38 0. 97 0. ,84 0. 90 1, .39 1. 02 0, .77 1. .16 0, .60 0. 69 1. 17 1. .17 12 .06 1. ,008 5. ,24 Penticton 1. .22 0. 89 • 0, .77 0. 73 1, .12 1. 35 0, .94 0. ,77 0, .78 0. 94 1. 08 1, .15 11 .74 1. ,140 4. .91 Summerland 1. .10 0. 84 0, .70 0. 68 1, .07 1. 35 0, .96 0. .80 0, .77 0. 96 1. 01 1, .21 11 .45 1. ,491 4. .86 Vernon 1. ,63 0. 96 0, .95 6. 99 1, .25 1. 56 1, .33 2. ,10 1, .36 0. 84 1. 44 2. .01 16 .42 • 1. ,700 7. ,23 Coldstream Ranch 1. .57 1. 19 0, .91 0. 73 1, .41 1. 30 1, .10 1. .15 1. .19 1. 38 1. 33 1. .75 15 .41 1. ,582 5, .69 Chase 1. .88 1. 39 0, .65 0. 91 1, .00 1. 48 0, .98 1. ,35 1. .40 0. 94 1. 38 1, .51 14 .87 1. ,165 5. .72 Falkland 2. ,01 1. 34 0, .96 1. 03 1, ,51 1. 26 1, .19 .1. ,69 1. ,60 1. 11 1. 67 1, .93 17 .30 1. ,500 6. ,68 Kamloops 1. ,32 0. 71 0. ,28 0. 36 0. ,59 1. 18 0, ,88 1. ,12 0. ,84 0. 58 0. 72 0. ,97 9.. 55 1. ,150 4. .13 Sicamous 2. ,92 2. 00 1. .31 1. 24 1, ,87 2. 71 1. .79 2. ,20 2. .28 1. 85 2. 45 2. .86 25 .48 1, ,400 9. .81 Westwold 1. ,27 1. 08 0, .78 0. 66 1, .13 1. 68 1, .11 1. ,19 1. ,11 •1. 18 1. 10 1. .46 13 .75 2. ,020 5. ,77 Source: Department of Transport, Gonzales Observatory, Victoria, B.C. TABLE VII. Precipitation and mean temperatures of selected sites i n 1971. Monthly Precipitation Monthly Mean Temperatures Mar. Ap r i l May June July Aug. Sept. Apr i l - . Mar. \"April May June July Aug. Sept. Aug. Oliver 0. 62 0. .36 0. 87 1. ,36 0, .71 0, .56 1. .12 3. ,86 40. ,1 46. ,8 59. 5 61. .3 68 .0 71. 6 55. 6 Penticton 0. 53 0, .65 1. 16 2. 39 0, .97 1, .32 0, .87- 6, .49 36. 6 47. ,1 57. 7 60, .4 68 .5 72. 4 56. 0 Summerland 0. 31 0, .18 1. 17 . 2. 06 0, .47 1, .19 0, .76 5. ,07 40. 0 45. ,9- 58. 2 59, .5 69 .8 74. 0 56. 6 Vernon 0. 57 0, .09 •1. 43 2. 64 0, .86 0, .93 0, .70 5, .95 38. 8 45. ,0 57. 2 59, .8 68 . 3 71. 5 55. 5 Coldstream 0. 54 0. .28 47 Ranch 2. 2. 41 1, .12 0. .69 . 0. .38 6. ,97 37. 3 43. •9 •55. 6 57. ,6 66 .6 71. 8 54. 1 Chase INC. 0, .02 1. 15 1. 57 0. .61 0, .54 0. .90 3. ,89 INC. 45. ,0 57. 9 . 59. ,7 67 .5 70. 3 55. 1 Falkland 1. 09 0. ,17 1. 53 2, 99 1. .48 1. .10 1, .56 7. ,27 36. 2 43. .3 55. 8 58. ,3 • 66 .0 69. 9 53. 3 Kamloops o. 42 0. .30 0. 95 1. 30 0. .64 0. .25 0. ,41 3. ,44 41. 2 47. 9 59. 0 62. ,3 72 .0 74. 2 58. 1 Sicamous 1. 02 0. ,34 2. 40 4. 80 1, .72 1. ,43 1. .61 10. ,69 37. 3 44. ,2 . 57. 4 59. ,9 68 .2 71. 7 54. 7 Westwold 0. 69 0. .57 1. 49 1. 33 0, .60 1. ,00 0. ,86 4. ,99 35. 5 40. 8 53. 3 56. ,2 64 .1 68. 6 50. 5 Source: Department of Transport, Gonzales Observatory, Victoria, B.C. 13 Okanagan- Kamloops region. The annual precipitation and the April-August precipitation ranges indicate that C. maculosa.is somewhat more tolerant to mesic, conditions. The presence of C_. maculosa i n a dense roadside infestation i n the relatively moist West Kootenay region exemplifies this point. The C_. maculosa infestation i n the arid region near Walhachin can partially be explained by the increased moisture regime along the pavement edge of the old roadbed. C. maculosa i n this area has not . spread substantially into the surrounding arid rangeland. The s o i l analysis (Table VIII) indicates that the Centaurea species readily colonize different soils'with a wide range of chemical properties. Plant density of C. diffusa and C. .maculosa.could only be significantly (P = 0.05) .correlated with the degree of disturbance of the s o i l . The greater the degree of disturbance, the higher was the plant density, of the knapweed species. These results indicate, as Atkinson and Brink (6) had earlier suggested, that any s o i l in the dry southern interior of the province of British Columbia with a disturbed A horizon i s subject to knapweed infestation. Marsden-Jones and Turrill- (55) indicated that the range of knapweeds i n the British Isles i s largely dependent on man-made or man-modified habitats. C_. diffusa.and C. maculosa infest a substantial range of ecological habitats i n the interior of British Columbia. Perhaps the apparent slight variation i n species distribution of the knapweeds can be attributed,to time of propagule movement into a particular area as both C. diffusa,and C. maculosa, are adapted to similar climatic situations. However, C. maculosa, apparently is, somewhat more adapted to the cooler more mesic TABLE VIII. Chemical analysis of knapweed infested soils ''Diffusa Sites \"Maculosa Sites D-l D-2 D-3 D-4 D-5 D-6 D-7 D-8 M-1 M-2 M-3 Plant density (/m2) 26.75 227.50 135.50 142.75 544.50 405.75 184.75 65.00 463.25 138.00 200.00 Disturbance Rating (1-3) 1 3 2 2 3 3 3 1 3 3 3 pH (H20) 7.18 6.85 6.89 6.79 8.35 6.93 7.18 5.91 6.41 7.38 7.03 pH (0.01M CaCl 2) 6.58 6.31 6.28 6.07 7.79 6.14 6.35 5.33 5.78 6.24 6.43 Conductivity (urn hos/cm2) 226.75 244.25 191.25 139.50 177.75 150.50 93.75 166.25 122.00 115.75 128.00 Sulfur (%) .065 .032 .031 .029 .021 .008 .020 .042 .042 .017 .014 Carbon (%) 3.47 1.42 2.91 0.95 0.88 0.51 1.43 4.04 3.64 1.34 1.86 Organic matter (% C-1.7) 5.90 2.41 4.95 1.62 1.49 0.87 2.44 6.87 6.19 2.27 3.15 Nitrogen (%) .319 .135 .238 .081 .061 .035 .117 .299 .329 .102 .117 C/N 10.99 10.50 12.35 11.87 13.98 14.98 12.35 13.80 11.50 14.22 16.51 Phosphorus (ppm) 2.05 1.45 2.25 1.83 1.85 1.83 2.23 5.43 2.20 2.80 3.03 Sodium (m.eq./lOOg) 0.858 0.098 0.083 0.138 0.220 0.265 0.180 0.130 0.120 0.075 0.165 Potassium (m.eq./lOOg) 3.670 1.670 1.185 0.983 2.468 2.825 1.688 2.155 1.733. 1.310 1.593 Magnesium (m.eq./lOOg) 12.110 2.015 1.608 1.703 3.458 1.265 1.843 3.470 2.608 2.860 4.33 Calcium (m.eq./lOOg) 11.935 6.345 10.153 4.548 16.030 4.273 7.560 10.200 10.153 9.655 11.000 Total Exchange capacity (m.eq./lOOg) 32.065 16.380 20.803 12.590 14.798 8.560 17.840 30.668 28.318 15.335 24.310 ''Sites: D-l Lac du Bois, D-2 Pritchard, D-3 Lavington, D-4 Summerland, D-5 Westbench, D-6 O.K. Falls. D-7 Myers Flat, D-8 Vernon, M-1 Chase, M-2 6-mile, M-3 Westwold. 15 conditions while C. diffusa readily colonizes semi-arid regions character-i s t i c of a large portion of the southern interior of the province of British Columbia. •16 I I I . .WEED ECOLOGY A. Literature Review • Centaurea diffusa and C. maculosa are members of the Cynareae tribe of the Compositae family. C_. diffusa i s described as an annual, biennial, or short-lived perennial. Vorobiov (94), studying the biology of two year old weeds of the Danube steppe, described C. diffusa as a biennial, remaining i n the form of a rosette, but sometimes flowering. The species i n British Columbia exhibits the biennial habit, but occasionally persists for a third season (triennial). Centaurea diffusa Lam. i s described as \"...pubescent becoming glabrate, rough, with an elongated taproot; Stems erect, 5-8 dm. t a l l , angled but not winged, branched near or above the base; Leaves alternate, the basal ones in a whorl, bipinnate to bipinnatifid, oblanceolate to oblong, up to 20 cm. long and 5 cm. wide, short-petioled, the ultimate segments narrowly oblong to e l l i p t i c , usually acute and wedge-shaped, the stem leaves sessile, the lower leaves bipinnate to bipinnatifid, the upper-most leaves bractlike and entire or minutely lobed; Flower heads solitary, more or less clustered at the ends of the branches, 1.5 cm. long; Involucre narrowly ovate or oblong, about 1 cm. long, woolly becoming glabrate and granular, the phyllaries leathery, nerved, the outer and middle phyllaries broadly to narrowly ovate, pale yellowish-green with a light-brown margin, the upper part narrowed into a s t i f f spine, the inner phyllaries lanceolate, tipped by a papery or leathery fringed appendage, spiny or spineless; Flowers white, pink or lavender, the outermost flowers sterile,- inconspicuous, with thread-like corolla lobes; Achene- oblong, 2.5 mm. .long, dark-brown, marked with several 17 conspicuous to faint, pale-brown or ivory lines; Pappus none, or on inner achenes as white chaff scales less than 1 mm. long\". (4). C. diffusa i s native, to the eastern and southern Balkan Peninsula, southern Russia, and western Asia (93). C. maculosa has also, been described as annual, biennial, or short-lived perennial. • The annual habit of C. maculosa has not been observed in British Columbia. The biennial habit and-more commonly the perennial habit have been observed with many plants being maintained for a period of at least four years. Centaurea maculosa Lam, i s described as \"... reproducing by seeds; Stems erect or ascending,, with slender wiry branches, rough-pubescent, 3-10 dm. high; Leaves alternate, pinnatifid, with narrow divisions, rough-divisions, rough-pubescent, the upper leaves often linear; Flower heads terminal and ax i l l a r y , numerous and clustered, 1.5-2.5 cm. in diameter, many-flowered; Involucre pale, 1-1.4 cm. high, i t s smooth strongly ribbed out and median ovate phyllaries with firm points and 5-7 pairs of c i l i a , the dark t i p 1-2 mm. long, the innermost phyllaries entire or fringed; Receptacle b r i s t l y , f l a t ; Flowers a l l tubular, whitish to pink or purple, perfect, the marginal flowers enlarged, falsely radiate, neutral; Achene brownish, about 2 mm. long, notched on one side of the base; Pappus with a short tuft of bristles at t i p end, 1-2 mwu long persistent\" (4). C. maculosa is native to Europe. The species' names characterize the difference between these two Centaurea species. Diffusa describes the diffuse or spreading habit of C. diffusa. Meanwhile, the name maculosa i s derived from \"macula\" meaning •' spot' which aptly describes the spotted appearance of the flower heads due to the terminal blackish fringe of the bracts of the flower 18 head of C_. maculosa. Hybridization between C. diffusa and C. maculosa has been regarded as a possibility because of variations which appear i n the populations of C. diffusa. These C_. diffusa variants appear similar to typical C. diffusa with r i g i d terminal spines, but the flower color tends more toward the purple of C. maculosa and the bracts have the terminal blackish fringe which i s a distinguishing characteristic of C_. maculosa. The achenes of these C. diffusa variants also exhibit a substantial pappus. Cytological studies by Moore and Frankton (60) indicate the chromosome complements of-C. diffusa as.diploid with 2n = 18 and C.. maculosa as tetraploid with 2n = 36. A hybrid between these two species would be expected to have the t r i p l o i d chromosome number or 2n•= 27. However, Moore and Frankton did not find any plants with the t r i p l o i d number indicating that hybrids did not occur between these two species i n the collections from British Columbia. Dr. R.L. Taylor (91) has confirmed these results with a more thorough.investigation of the Centaurea populations i n British Columbia. Taylor suggests that the degree of variation within the C. diffusa population i s possible due to more than one introduction of the species into the area. He also . suggests that the variable genotypes expressed by flower color i n C.- diffusa may be due to loose multiple gene control. Preliminary studies by Taylor indicate that C_. diffusa and C. maculosa have two distinct pigment systems which are responsible for the development of the variable flower color. Saarisalo-Taubert (77), studying hybridization of Centaurea, section Jacea, i n Finland, found hybrids and hybrid swarms between 19 members of the Jacea group of Centaurea common wherever the ranges of different species overlap. Hybrid plants (C. jacea x C_. nigra) or pure C. nigra were introduced into Finland but strong selection in favor of • the jacea genes has resulted i n disappearance of pure C. nigra, popula-tions (77). Simmonds (85) included C. diffusa and C. maculosa as non-toxic weed species of veterinary importance. The sharp spines, characteristic of many Centaurea species, can cause mechanical injury to the mouth and digestive tract of grazing animals resulting i n considerable losses i n beef production. Popova (67) indicated, even though C. diffusa was found to contain 0.07% alkaloids and 0.75% glucosoide, the harmful effect of diffuse knapweed i s s t r i c t l y mechanical i n nature and no cases of poisoning have been recorded i n Crimea. The spiny nature of the mature diffuse knapweed plants causes considerable discomfort .to humans. Renney (71) pointed out that diffuse knapweed presents a dense barrier i n many recreational areas i n the southern interior of the province of British Columbia. A number of Centaurea species, notably C_. repens and C. s o l s t i t i a l i s , have been shown to be the causal agents of the disease, nigropallidal encephalomalocia, i n horses (103,104). Horses exposed to nutritionally adequate rations (grass hay supplement i n addition to the weed hay being tested) suffered from the disease. Nigropallidal encephalomalocia has been reported only i n horses exposed to Centaurea species. Young et a l . (103) suggested that any toxic principle that might be present i n the plant i s not accumulated or retained i n the tissues of the horse which • 20 feeds on the weeds intermittently. . Centaurea americanai'^, Nutt. has been characterized as a cyano-genetic plant producing considerable amounts of hydrocyanic acid (63). Further studies of the Centaurea genus w i l l probably yield.other toxic compounds of importance. Popova (68) has indicated the nutritive value of diffuse knapweed i n the early,stages of growth' (Table IX). Centaurea. species i n the flowering stage of development are very high i n fibre content (Tables IX and X) and are not u t i l i z e d as forage. Popova (68) suggested that in the winter, the leafless stems of C. diffusa become moistened and can again be eaten, but the mature stalks contain l i t t l e nutritive value (Table IX). Several Centaurea species are recognized as honey plants and C_. nigra, C_. jacea, and C_. cyanus are commonly u t i l i z e d as sources of nectar i n North America and Europe. Honey produced from C. maculosa i n Virginia and Pennsylvania i s described as light i n color, mild flavored and of good quality -(53,54). The spotted knapweed i s generally thought of as a weed i n these areas but,, as one bee keeper explained: \"the plant i s not very prickly and has a most attractive appearance. Although i t i s a weed, i t i s one which could be tolerated even i n a flower garden\" (53). Maurizio and Grafl (56) reported that C_. cyanus produced approximately 5 mg. of nectar per day per flower head. Bees collected approximately 28 mg. pollen per plant from C. montana (56). However, pollen analysis showed only 1-4% of the sample was that of C. cyanus while clover and rape made up 96-99% of the sample (56). TABLE IX. Chemical Composition of the Above Ground Parts of Centaurea diffusa.(% dry wt.) Plant Portion Date . Collected Ash Crude Protein True Protein Fat Cellulose N-Free Extrad Spring Rosette 12/6/55 8.43 11.26 8.71 3.26 .33.62 40.31 F a l l Rosette 7/9/55 7.12 10.23. 7.93 3.12 34.17 40.86 Bolting 14/5/55 7.29. 10.78 7.81 3.13 33.71 41.67 Budding 26/5/55 6.94 9.56 7.12 .2.94 . 34.31 42.13 Flowering 18/6/55 . . 6.24 9.42- 7.00 3.22 - 40.52 39.46 Fruiting 21/9/55 6.14 6.54 6.94 3.18 47.24 35.12 Autumn Hay- 17/9/55 6.12 6.42 5.42 0 44.32 0 Winter Hay 10/1/56 3.81 5.31 0 0 ' 48.63 0 Source: Popova (1964) TABLE X. Chemical Analysis of C. diffusa and C. maculosa i n Flower (% dry wt.) (a) C. diffusa Moisture Ash Protein Nitrogen Fat Crude Fibre' •N-Free .Extract Root 3.51 • 13.73 3.71 . 0.59 1.19 29.40 • 51.97 Stem 4.52 4.07 ; 2.74 0.43 0.80 41.80 50.59 Leaves 2.43 15.62 8.31 1.33 2.34 21.75 51.98-(b) C. maculosa Root 1.37 43.60 2.62 0.42 0.78 31.10 -Stem 2.86 6.53 5.40 • 0.87 1.23 43.55 43.29 Leaves 3.73 16.18 5.68 0.92 2.87 16.85 46.42 Source: Fletcher (1961) 23 Knapweed seeds may form an important part of the diet of some birds. Birds may cause considerable damage to ripe or nearly ripe seeds. Marsden-Jones and T u r r i l l (55) observed that goldfinches were extremely fond of knapweed seeds i n Great Britain. Coxworth et a l . (15) have explored .the possibilities of u t i l i z i n g dry land weeds as potential high protein seed crops for semiarid areas. Working with three Chenopodiaceae species which produced high seed-protein yield and drought resistance, Coxworth et a l . found kochia and Russian t h i s t l e had comparable protein yields to flaxseed or rapeseed. Earle and Jones (18) suggested .that the Compositae family was one of the families which appears to offer outstanding promise as potential sources of new oilseeds. The u t i l i z a t i o n of range weeds such as Russian t h i s t l e or diffuse knapweed as high protein seed crops for semiarid areas may be of some importance i n the future as world protein shortage continues to increase. Besides the u t i l i z a t i o n of plant species as sources of high-protein seed meals considerable interest and study i s devoted, to the search for chemical compounds such as alkaloids which may be of indus-t r i a l or medical use. Earle and Jones (18) found that C_. cyanus• and C. americana were alkaloid positive species.,, Hultin and Torsell (45) also found alkaloids present i n plant parts of Centaurea species. Other workers have found a vast array of chemical compounds i n numerous plant species. • The Compositae family offers considerable potential i n the discovery of new, useful chemical compounds of natural plant origin. In certain parts of Europe C. diffusa i s being u t i l i z e d as a • medicinal plant. Racz et-al. (69) suggested that i t was feasible to . 24 u t i l i z e a number of Centaurea species, including C_. diffusa, as amara. Amara, or the Spanish word \"amargos\", i s described as a vegetable drug with a bitter taste commonly referred to as \"bitters\". The medicine i s thought to stimulate the appetite and to favor digestion. Racz et a l . determined the bitter principle organoleptically i n natural species (including C. diffusa) and found the b i t t e r value of the aerial part i n the concentration of 1:2000-3000. An interesting aspect of the biology of C. maculosa has been described by Cavallit'o and Bailey (13) and more recently by Monya et a l . (58) as the isolation of an antibacterial substance from the dry leaves of this species .• Cavallito and Bailey found that C_. maculosa yields approximately 1.5% of an unsaturated lactone antibiotic, C^g^gO^, from the dry leaves. It inhibits the growth of both gram positive,and gram negative bacteria. Monya et a l . found the antibacterial principle of C. maculosa to inhibit the development of Salmonella typhi and Shigella shiga. The C.: maculosa antibiotic Is not being u t i l i z e d , but i t may be useful i n the future since i t induces l i t t l e or no • development of resistant strains of bacteria. Popova (67) indicated the C. diffusa seeds have germination rates of 90-100% and this rate i s maintained under various conditions. Atkinson and Brink (6) found that germination of C. diffusa occurred in temperatures of 10°C to 25°C with 15°C to 20°C being optimum. They found germination rates i n the order of 73% for current year seed and 74% for year old seed. Zednai (105) found C. maculosa seed exhibited 57.4% germination immediately after maturity and 86% to 100% germination after;3 months... ;_v His studies revealed that germination of 25 C. maculosa seed was reduced to 67% after storage for one year at room temperature. The knapweeds are entomophilous. The insect pollinators are seeking pollen and/or nectar. Marsden-Jones and T u r r i l l (55) pointed out that insects favor cross-pollination with pollination of the British knapweeds normally brought about by nectar seeking insects. They found that an insect pushes down i t s proboscis successively into many florets with open corollas. Each floret i s probed separately whether for pollen or nectar and a single insect can pollinate many florets without flying. Each floret has only one ovule requiring only one viable and compatible pollen grain for f e r t i l i z a t i o n . Additional florets besides the one being probed may be f e r t i l i z e d by the insect. Marsden-Jones and T u r r i l l observed that self-pollination by. recurving of stylar arms or stigmatic lobes does not occur, but 'own' pollen grains may be deposited on stigmatic surfaces since the florets i n a capitulum develop acropetally over a period 1 to 2 weeks. Marsden-Jones and T u r r i l l suggest that most knapweed species exhibit self-incompati-b i l i t y . However, Taylor (91) has found that C. diffusa and C. maculosa are both self-compatible. Marsden-Jones and T u r r i l l suggested that reproductive capacity of the knapweeds depends upon,'', number of reproductive seasons, number-of capitula per plant, number of potentially f e r t i l e florets, effective-ness of pollination and subsequent f e r t i l i z a t i o n , number of cypselas produced, survival of cypselas, percent germination of the seeds, and establishment of seedlings. 26 Gland-tipped hairs and sessile glands were observed on the surface of leaves of the knapweed species by Ormrod (64). Ormrod and Renney (65) suggested that the development of trichomes on the surface of C. diffusa leaves results i n a pubescence sufficient to influence retention or penetration of herbicidal sprays. The a b i l i t y of Centaurea species to readily invade disturbed sites and withstand competition from other plant species has been recognized for some time. The knapweeds appear to exhibit allelopathic effects on other plant species. Fletcher and Renney (27) confirmed the presence of an inhibitory substance i n C.. repens, C. maculosa and C. diffusa with the leaves containing the highest portion of the inhibitor (Tables. XI and XII). Field observations indicate seedlings of the same species develop i n close proximity to the parent plant partially ruling out self-toxicity. Fletcher-and Renney part i a l l y characterized the inhibitor as an indole derivative or as a precursor of auxin. These studies explain, i n part at least, the aggressive nature of the Centaurea species which enable them to rapidly colonize disturbed sites and withstand the competition from associated plant species. TABLE XI. Effect of C. maculosa Water Extract on Crop Seedlings (7 days after sowing) Crop Extract Germ • (%) Shoot (cm) Root (cm) Lettuce Control H20 96 3.4 3.5 Leaf 80 2.6 1.5 Seed 80 2.0 • 1.9 Root 92 3.2 3.2 Barley Control H20 84 12.6 10.2 Leaf 72 7.8 6.5 Seed 84 3.5 . 3.6 Root 84 . 12.1 9.8 Source: Fletcher (1961) TABLE XII. Effect of C. diffusa Water Extract on Crop Seedlings (7 days after sowing) Crop Extract Germ (%) \" Shoot (cm) Root (cm) Lettuce Control H20 96 3.4 3..5 Leaf 84 2.6 1.0 Root 84 3.1 3.5 Barley Control H20 84 12.6 10.2 Leaf 72 7.7 6.4 Root 80 11.4 9.1 Source: Fletcher (1961) 28 B. Experimental Methods Study areas were established at 13 sites i n the Okanagan-Kamloops area. The 8 C.. diffusa and 5 C. maculosa sites were representative • knapweed infested areas in the interior of B.C. Five Centaurea plants within 1 meter square quadrats were choosen at random at each of the 13 sites. Phenological and developmental data were recorded on these plants throughout the 1971 growing season. Capitula were collected at random at each of the study sites and the number of seeds per head was recorded. . Line transects were established through knapweed infestations, and random one meter square quadrats were harvested to include a l l plant growth. The forage species were separated from the knapweed and the samples were dried at 50°C for a period of 24 hours. , Field observations'were recorded from knapweed infested areas throughout the southern interior of British Columbia during the summers of.1968, 1970 and 1971. «• 29 Laboratory Studies (a) Preliminary germination studies Germination studies were conducted on C„diffusa seed collected in the. spring of 1971 from flower heads which had retained a few seeds over the previous winter and on C. diffusa and C. maculosa seed collected i n the late summer of 1970 and stored at room temperature for 20 months, C. diffusa and C. maculosa seed collected at Chase i n the late summer of 1971 and stored for 3 days and 25 days at room temperature. Each test included two replicates of 50 seeds placed on moistened, sterilized No. 1 Whatman f i l t e r paper i n glass petri dishes. Dishes were placed in a Seedburo germinator at 25°C. Percent germination was determined after one week. (b) Effect of light on seed germination Studies on the effect of light on seed germination were conducted in two Percival growth chambers with the temperature maintained at 25°C (± 1°C) and with controlled light conditions. A l l treatments were con-ducted by placing 25 seeds, of one of the knapweed species on moistened sterilized No. 1 Whatman f i l t e r paper i n glass petri dishes. Seeds of C. diffusa and C_. maculosa were harvested i n the late summer of 1971 and stored at room temperature u n t i l being u t i l i z e d i n the experiment. The light treatments included continuous light of 700 f t - c in the f i r s t chamber, and 16 hr. light of 700 f t - c followed by 8 hr. darkness in the second chamber. Continuous darkness was maintained by wrapping glass petri dishes i n aluminum f o i l and placing them i n the f i r s t chamber. The whole experiment was conducted three different times with the light 30 conditions being altered between the two chambers. Three replicates of each species i n each treatment occurred i n a l l three runs. Percent germination-was recorded after two, four, and six day periods. Seeds i n the dark were removed and a new set of seeds was ut i l i z e d for each period. Data were analyzed by the analysis of variance and by, simple linear multiple regression. (c) Effect of temperature on seed germination A linear temperature gradient bar was constructed after Timbers and Hocking (92) to study the effect of temperature on germination of C. diffusa and C_. maculosa seed. Under operating conditions temperature d r i f t at any location along the bar was less than ± 0.5°C. Temperatures along the bar at which seeds were placed were 7°C, 10°C, 13°C, 16°C, 19°C, 22°C, 25°C, 28°C, 31°C and 34°C. The bar was sectioned off into ten compartments with insulation and each compartment was equipped with individual l i d s . The•removable l i d further insulated the bar and maintained dark conditions (Figures 5 and 6). A l l seeds of C. diffusa and C. maculosa u t i l i z e d i n the study were collected i n the late summer of 1971 and prior to the test were treated with 1:6 sodium hypochlorite solution to control fungi. Seven seeds of each species were placed i n a row, perpendicular to the length of the bar i n each of the compart-ments on moistened, s t e r i l i z e d No. 1 Whatman f i l t e r paper. The seeds were exposed to diffuse light for less than 30 seconds daily as percent germination was recorded. Water was added daily to maintain moistened condition at the warm end of the bar. The experiment was conducted three different times. The test was discontinued after two weeks. 32 However, no additional germination occurred after 10 days. Data were analyzed by the analysis of variance and by simple : ^ curvilinear multiple regression. Greenhouse Studies • Studies were conducted under greenhouse conditions to determine the effect of seeding depth on percent germination and subsequent seedling emergence. Twenty-five seeds' of C_. diffusa were sown at depths of 0 cm., 0.5 cm., 1 cm., 3 cm., and 5 cm. i n five different 5\" plastic pots f i l l e d with greenhouse s o i l . The texture of the s o i l was determined as sandy, loam by the hand texture method. The treatments were replicated four times' and seeds of C. maculosa were treated i n a similar manner. The pots were placed i n a randomized complete block design. The temperature i n the greenhouse was maintained at 45°C (±5°C). The amount of water held by.the greenhouse s o i l at f i e l d capacity, or 1/3-bar water tension, was determined and the soils were brought to f i e l d capacity daily by the addition of sufficient water to reach the required weight for each pot. Cumulative percent emergence and mortality of emerged seedlings were recorded daily. Data were-analyzed by the analysis of variance and by simple linear multiple regression. 33 C. Results. and Discussion.... The Centaurea.species with their rosette habit of growth, generally escape grazing. However, i n the spring the plants bolt and are susceptible to grazing, but as the plants mature they become very woody and ,are essentially inedible. Cattle have been observed grazing ;C. diffusa-quite,extensively on some of the ranges i n the Pritchard area near Kamloops. This knapweed was i n the bolted stage. No adverse effects have been observed or reported i n these cattle populations.As the area i s heavily overgrazed and the cattle are forced to eat the knapweed they may acquire a \"taste\" for the bitter knapweed. Centaurea species are generally not u t i l i z e d as forage species.,, and, in addition, interfere with the consumption of more desirable forage species below the dense, spiny, overstory of the knapweed infestations. Knapweed infested rangelands have considerably reduced forage -yields. Selective grazing and the aggressive nature of the knapweeds tend to crowd out many palatable forage species. The large vegetative mass of the Centaurea species u t i l i z e substantial moisture and nutrients from the s o i l decreasing the potential forage production of other species on knapweed infested ranges.. Table XIII indicates the dry matter yields (kg/hectare) of infested and non-infested rangeland. These samples, were obtained from different rangeland conditions.. The data collected from the Summerland site indicate - the a b i l i t y of C. diffusa to reduce forage yield on native range which i s not ut i l i z e d by grazing animals, however, •34 TABLE XIII. Yield of forage and knapweed from knapweed infested rangeland i n the southern interior of British Columbia (Kg./he) Lac du Bois Summerland Commonage - Chase Diffuse Forage • Diffuse Forage - Diffuse Forage Spotted \"' Forage 0 1450 0 - 880 0 970 990 20 0 1440 0 920 0 940 , 1220 30 0 1150 0 ' 590' • 0 ' 630 1760 10 0 300 0 470 40 320 140 710 ,0 440 130 170 150 530 o 350 250 540 220 820 10 400 340 260 . 370 310 60 ,980 750 . 840 390 1100 150 400 820 750 660 210 . 290 • 530 850 220 770 20 * 0 1630 810 20 * 0 1700 820 70 * o 2060 890 20 R^andom samples from range reseeded to crested wheatgrass 35 a portion of the area was severely disturbed by a pipeline right-of-way. The decrease i n forage y i e l d was significantly (5% level) correlated with increased production of C. diffusa. The two other diffusa, sites, Lac du Bois and Commonage, were located on rangeland which was u t i l i z e d by domestic stock and indicate the problems of overgrazing with sub-, stantially reduced yields i n the knapweed-free plots. Occasionally, relatively large forage yields were observed when the knapweed yield was relatively large, indicating that' the forage species may be present, but due to the dense, spiny overstory of C_. diffusa, cattle were unable to u t i l i z e the available forage. C. maculosa also reduces the forage yield to a point where the rangeland i s not providing any available forage to the domestic grazing animal. A well managed, reseeded dryland range may produce approximately 2000 kg. of forage per hectare as compared with 400 to 500 kg. of forage per hectare on native' range (Table XIII). Infestation of native range-lands with Centaurea species results i n considerable reduction i n actual y i e l d and i n available forage. Heavily infested rangelands are essentially non-productive. McLean and Marchand (57) indicated that on the average 660 l b . of available forage i s needed for a 1000-pound cow for 1 month (animal-unit month). Stocking rates can be estimated on the basis of acres per animal-unit month (AUM) from yield data assuming a 45% carryover. They suggested that \"bluebunch wheatgrass - rough fescue\" range in \"good\" condition producing on the average 800 lb. forage per acre would have an average stocking rate of 1.5. Knapweed infested range may produce 36 only 100 lb. of forage per acre resulting in a stocking rate of 12.0 acre per animal-unit month. Losses i n potential w i l d l i f e production are d i f f i c u l t to compute, but i t could be speculated that a similar reduction i n the deer popula-tion to that of beef'production would occur on rangeland heavily -infested with 'Centaurea species.' Dryland hay i s often infested with C. diffusa, or C. maculosa i n the interior of B.C.. Knapweed infested hay i s generally not palatable to livestock resulting i n considerable losses. Popova (67) indicated that a l f a l f a fields i n the- Crimea are heavily infested with C. diffusa with the knapweed comprising from 40-65% of the dry matter produced. The production of dryland hay i s limited i n the interior of British • Columbia and C.. diffusa and C. maculosa are generally not a problem i n well managed irrigated hay fields. Wodehouse (100) suggested that the Centaurea species, which are insect pollinated and exceptionally well adapted to this mode of pollination, are unlikely to cause hayfever. Therefore, work loss due to hayfever allergies cannot, be attributed to the presence of Centaurea species. However, the knapweeds can cause a mild discomfort i f tasted which'can be u t i l i z e d as a useful diagnostic technique. I t i s d i f f i c u l t to place sale values on rangeland infested with knapweed, as compared to those ranges free of knapweed. However, the data i n Table XEIIindicates the substantial losses i n forage y i e l d and subsequent reduction i n beef production on rangelands heavily infested with knapweed species. The sale value of Centaurea infested rangeland 37 would undoubtedly be reduced to a low level. The a b i l i t y of the Centaurea spp. to establish readily i n disturbed environments indicates their, usefulness as pioneer species. The rapid establishment of plant cover i n the form of rosettes on barren soils prevents s o i l erosion. This plant cover also aids in the build up of organic matter i n the s o i l . Bees were observed in the Okanagan to be extensively working C. diffusa, C. maculosa and C. nigra. Bees were collected and nectar samples were withdrawn. These samples were very bitter to the taste. Samples of pollen collected by foraging bees were retained. The accom-panying photographs (Figure 7) of the pollen were taken of the three Centaurea species u t i l i z e d by foraging bees in the Okanagan (courtesy of. J. Louveaux (52)). It i s apparent that C. diffusa and C. maculosa provide substantial pollen and nectar in the interior of British Columbia. These plants may be considered as valuable honey plants in waste ground, but they do not compare to the established honey plants such as clover and a l f a l f a . The knapweeds cannot be regarded as having significant forage value due to their lack of palatability and high fibre content. Popova (68) indicates the nutritive value of knapweed hay, but suggests that C. diffusa may -be u t i l i z e d only i n the early bolting stage. The rosette habit of knapweeds enables the plants to escape grazing and the mature flowering -stalks are much too coarse with a very high fibre content. The disadvantages of these plant species heavily over-ride the potential value they may have as forage. Generally, the knapweed plants with.their tough, coarse stems and spiny flower heads are avoided 38 Coupe e q u a t o r i a l * v u e p o l a i r e - O r n e m e n t a t i o n Coupe m e r i d i e r m e B C V L Fig. 7. Pollen grains of C. nigra, C. maculosa and C. d i f f u s a * A - C_. nigra; B - C_. maculosa; C - C. d i f f u s a Source: Louveaux, J. (1971) t i l ? * Fig. 8. C. maculosa (upper) and C. d i f f u s a seeds 39 by grazing animals and are not particularly palatable to the livestock-and w i l d l i f e . No actual observations of birds or rodents u t i l i z i n g knapweed seeds i n the interior of the province were made. However, numerous neat piles of C_. maculosa chaff were observed at Chase indicating possible rodent or bird u t i l i z a t i o n of the seed. As described earlier, C. maculosa seeds are larger than C. diffusa (Figure 8). C. maculosa achenes have a substantial pappus whereas a pappus i s generally absent .from C_. diffusa achenes. The average weight of C_. diffusa seed i s 1.099 mg. with C_. maculosa seed weighing 1.778 mg. (Table XIV). Both of these species f a l l within the range of 'relatively large seeds' as described by Frenkel (29). TABLE XIV. Weight of C. diffusa, and C_. maculosa, seed gathered i n British Columbia C. diffusa . C. maculosa Sample . wt. 100 seeds wt. of single' (gm.) seed (gm.) wt. 100 seeds wt. of single (gm.) seed (gm.) 1 0.1096 • 0.001096 0.1849 0.001849 2 0.1079 0.001079 0.1762. 0.001762 3 0.1109 0.001109 0.1702 0.001702 • 4 0.1126 0.001126 0.1788 0.001788 5 0.1089 0.001089 0.1789 0.001789 Mean wt. 0.001099 0.001778 The seeds of Centaurea species germinate readily over a broad range of environmental conditions. Marsden-Jones and T u r r i l l (55) describe the germination process of C. nigra as follows: \"The cypselas 40 are slightly compressed and at germination a longitudinal s p l i t . appears in the pericarp, developing from below upwards. This s p l i t occurs at one margin i n the plane of .'compression.... The radicle grows through the s p l i t pericarp at the base of the cypsela\". Cypselas of C_. diffusa and C_. maculosa were observed to germinate i n a similar manner. C. diffusa which was retained i n the flower head' over winter exhibited 88% germination. . Seed of C. diffusa and C. maculosa exhibited 95% and 93% germination respectively after 20 months i n storage at room temperature. Three day old seeds of C. diffusa and C. maculosa exhibited 40% and 20%(germmat^r^ , 68%~and_80%>germination'respectively. TABLE XV. The effect of light on percent germination of C. diffusa and C. maculosa seed (adjusted means) Species Treatment Treatment Duration (Days) Two Four Six Light 58c* 67b 72b Diffusa Dark 69b 83a 87a Light + Dark 69b 83a 87a Light 58c 67b 72b Maculosa Dark 82a 83a 87a Light + Dark 69b 83a 8.7a \"Means for the same variable not sharing the same letter were s i g n i f i ^ cantly different at the 5% level according to Simple Multiple Regression. The results irr-.Table XV indicate that the continuous light treatment significantly (P = 0.05) reduced the percent germination of both Centaurea species after two, four and six days. The dark and the 16 hr. light plus 41 8 hr. dark treatments were not significantly different at the .5%.level after four and six days for both species. However, after two days, germination of C. maculosa i n the dark was significantly higher than C_. diffusa.in the dark and significantly higher than both species i n the light plus dark treatment. After four and six days the germination of C. maculosa i n the dark was no longer significantly higher than that of C. diffusa i n the dark, or both species i n the light plus dark treat-ment. C_. diffusa and C_. maculosa germinating under natural conditions • would not be affected by, light duration. The results shown i n Appendix Table I and Figure 9 indicates that germination of C_.- diffusa .and C_. maculosa occurs in the range of 7°C to 34°C. Germination of over 80% occurs i n the range of 10°C to 28°C for C. maculosa and in.the range of 13°C to 28°C for C_. diffusa. Analysis indicated that- after two days i n the temperature treatments C. maculosa had a significantly higher (P = 0.05) percent germination than C_. diffusa at temperatures of 13°C and 16°C.. Germination of both species appeared to be delayed at the cooler temperatures with C_. maculosa showing higher cumulative percent •germination at the lower temperature C7°-10O) than C. diffusa at the end of the experiment. C_. diffusa had a higher percent germination at the warmer temperature (34°C). The optimum temperatures were calculated from the f i t t e d curves from the regression analysis (Table XVI). These results suggest that C. maculosa w i l l germinate at slightly, lower temperatures. than C_. diffusa partially explaining the more northern distribution of C. maculosa. 42 Fig. 9. The effect of temperature on germination of C. diffusa and C. maculosa seed A) C_. diffusa - two days after sowing B) C. maculosa -two days after sowing C) £. diffusa - ten days after sowing D) C_. maculosa -ten days after sowing TABIJB XVI. Optimum germination temperatures for 143 C. diffusa and C. maculosa Time after Optimum temperatures (°C) sowing (days) Maculosa Diffusa 1 22.9°C 25.5°C 2 21.6 23.0 3 20.8 21.8 4 20.3 21.0 • 5 19.6 20.6 6 19.4 20.6 7 19.4 20.5 8 19.0 20.5 9 19.0 20.5 10 19.0 20.5 Fig. 10. Seedling emergence of C. diffusa and C. maculosa 24 days after sowing at different s o i l depths. 0 0.5 1 3 5 Depth of sowing (cm.) V * M - C. maculosa; D - G. diffusa 44 Twenty-four days after sowing C_. maculosa exhibited a significantly (P-= 0.05) larger cumulative percent emergence than C_. diffusa at a l l sowing depths (Figure 10 and Appendix Table I I ) . The optimum sowing depth for each species was at the s o i l surface. Emergences,from sowing depths of 0.5 cm. and 1 cm. were not. significantly different for either species. There was a significant (P = 0.05) decrease i n percent emer-gence as depth was increased from 0 cm. to 0.5 or 1 cm. , from 0.5 or 1 cm. to 3 cm. and from 3 cm. to 5 cm. C. maculosa seed i s capable of emer-gence at the 5 cm depths. C. diffusa did not emerge from depths below 3 cm. which confirms Popova's (67) findings. The larger seed of C. maculosa apparently enables this species to establish seedlings -from greater s o i l depths. Seedling mortality was recorded during the above emergence study. Under greenhouse conditions 7.6% and 4.0% of the C. diffusa and C. maculosa seedlings, respectively, died. Mortality was primarily due to damping-off fungi and occasionally to the lack of chlorophyll. Seedling mortality i n the f i e l d was observed to be over 55% i n some instances, but generally appeared to be approximately 12%. Further studies are warranted to determine the actual f i e l d mortality of seedlings of the' Centaurea species i n the interior of British Columbia. There does not appear to be any morphological difference between seedlings of C. diffusa and C. maculosa, l i t t l e variation being evident between the cotyledons and juvenile leaves of the two species. The f i r s t true leaves were observed under greenhouse conditions approximately two weeks, after emergence. 45 Both species 'form rosettes which are the over-wintering stage. Morphological differences between rosettes of C. diffusa.and C. maculosa are uncommon. The rosettes generally consist of 10 leaves but may range from 6 to 28 individual leaves. The much divided leaves are on the average 6.6 to 7.1 cm. long. Rosette mortality i s rare. The rosettes of C. diffusa appear to require a cold period to induce transition to flowering. Thus seedlings established i n the spring were not observed to complete their l i f e cycle and produce viable seed i n their f i r s t season of growth. Greenhouse studies indicated the a b i l i t y of C_. maculosa to produce viable seed i n one season of growth. Transition to flowering (bolting) of the knapweed species occurs in early May i n the interior of B.C.. C. diffusa rosettes produce only one stem but occasionally two stems have been observed. C_. maculosa rosettes produce one to six stems and perennial plants have been observed to produce 15 or more stems. Stem height of C_. diffusa plants i s from 10 to 20 dm. whereas C. maculosa plants are .20 to 30 dm. t a l l . Formation of the numerous reproductive flower buds can be observed in early June. Flowering occurs i n July and August with i n i t i a l flowering of C. maculosa being observed i n the f i r s t week of July with C. diffusa flowering approximately two weeks later. Mature seeds were formed by mid-August. The data i n Table XVII indicates the average seed production per-plant of C_. diffusa and C. maculosa under semiarid rangeland conditions in the interior of B.C. . Assurrdng 80% germination, the annual reproduc-tive capacities of C. diffusa and- C... maculosa are 665 and 298 respectively. 46 Reproductive capacity was also determined for the knapweed species growing under irrigated conditions at the Canada Department of Agriculture, Research Station, Kamloops, B.C. TABLE XVII. Average seed production of C. diffusa and C. maculosa Flower heads per plant Seeds per head Seeds per plant Annual Reproductive Capacity C. diffusa 66.5 12.5 832.3 665 . C. maculosa 14.0 26.6 372.4 298 irrigated C. diffusa • 1403.7 13.0 18,248.1 14,599 irrigated C. maculosa 706.7 35.8 25,299.9 20,240 Vegetative reproduction does not occur naturally i n C. diffusa populations. C. maculosa does exhibit vegetative spread. A number of lateral shoots arise just below the s o i l surface and grow horizontally for approximately 3 cm. then form a rosette. These rosettes generally flower the following season but do not normally become detached from the parent rootstock. . This form of vegetative reproduction confers the perennial habit observed i n many C. maculosa populations. The triennial habit observed i n some C_. diffusa populations arises from regrowth of parent rootstock and not from the- development of lateral shoots. Seed dispersal of C. diffusa i s mainly by wind as mature plants often break off at ground level and function as tumbleweeds (67,71). 47 Atkinson and Brink (6) pointed out that dispersal close, to the parent plant is facilitated by horizontally placed involucres, which open as dehydration occurs, dropping their achenes readily.- Renney (71) found that C. maculosa populations extend largely through peripheral enlarge-ment' of the existing stand. C_. maculosa exhibits a 'flicking' action which spreads seeds up to several feet from the parent plant. Marsden-Jones and Turrill. (55) indicated that the presence of a pappus on the achene has l i t t l e or no effect in enhancing•dispersal efficiency. Field observations suggest that both C_. diffusa and C. maculosa have dual mechanisms of seed dispersal. The dual seed dispersal enables these species to rapidly colonize the vicinity near the parent plant and to establish new populations some distance removed. The 'flicking', action of C. maculosa 'suggests a spring-like action. The apparent ability of the C_. maculosa achenes to be propelled from the receptacle is due tp dehydration ...The bracts of the flower heads open up enabling the loosely held achenes to be expelled from the head i f movement of the plant stems occurs (Figure I ' D . \" Agents such as insects, wind,, or animals cause sufficient movement and .subsequent release of the achenes. The released C. maculosa achenes, with their substantial pappus, may become attached to passing animals and. transported some distance from the parent plant. Both knapweed species are common along roadsides in' the dry interior .and well adapted to 'seed dispersal along routes of travel. C. diffusa plants readily become attached to vehicles and achenes are individually dispensed through-, the relatively small distal opening in the receptacle (Figure 11). C_. diffusa and C. maculosa achenes may also be transported by mud adhering to vehicles. The pappus of 48 C. maculosa achenes does not f a c i l i t a t e wind dispersal. The normal development of C_. diffusa and C. maculosa plants involves the production of entire juvenile leaves followed by the production of the more typical much divided adult leaves. The rosette habit i s lost when the plant undergoes transition to flowering and the formation of lobed-leaves ceases with the formation again of smaller, entire leaves. The development on one plant of leaves of various shapes i s termed heteroblastic development. Feldman and Cutter (22,23) studying the regulation of leaf form in Centaurea s o l s t i t i a l i s L. have evidence suggesting that endogenous gibberellins affect the type of leaf produced. They speculate that when the seed germinates the first-formed, simple, entire leaves are partly a response to high levels of endogenous gibberellins. The commencement of lobed-leaf formation may occur due to GA levels being reduced. When plants undergo the transition to flowering, endogenous gibberellin levels rise again as reflected by the production of small entire leaves. Feldman and Cutter propose \"...a gibberellin.or gibberellin-like system operates in the whole plant to modify or regulate the type of leaves produced i n the heteroblastic series\". The commonly associated plant species observed i n the knapweed infested areas of the interior of the province of British Columbia included: Achillea lanulosa, Agropyron spicatum, Artemisia frigida, Balsamorhiza sagittata, Bromus tectorum', Festuca idahoensis, .F. scabrella, Koeleria cristata, Lupinus sericeus, Poa pratensis, P. secunda, Stipa comata, Taraxacum officinale, Lappula echinata, Tragopogon pratensis, Gypsophila paniculata, Verbascum thapsis, Potentilla spp. , Rumex acetosella,'Chrysothamnus nauseosus,. Astragalus spp. , Antennaria umbrinella. Forage species such as bluebunch wheatgrass and rough fescue are being replaced by other, less desirable, grass species such as Sandberg's bluegrass and downy brome and weedy species such as knapweeds and sages on most of the ranges i n the interior of British Columbia. 50 Fig. 11. C . maculosa ( l e f t ) and C . d i f f u s a flower heads 51 IV. CONTROL A. Literature Review Knapweeds do not generally present problems on cultivated land. Therefore, on land which i s plowable, control of C. diffusa and C. maculosa can be readily achieved. Atkinson and Brink (6) pointed out that regeneration from \"plowed-under crowns\" was not frequent, but incompletely covered crowns regenerated readily. Popova (67 indicated that as a result of. shallow plowing knapweed seeds which had been more than 3 cm. under the surface were brought to the surface and i n subsequent years knapweed flourished on these plots (see Table XVIII). However, TABLE XVIII. Shallow plowing (7 cm.) on C. diffusa i n the fruiting stage (% cover) Control 1955 .1956 1957 1958 Just before Treatment 1955 Treatment 1956 1957 1958 Grasses 6.0 4.5 3.6 4.1 6.3 4.5 3.9 3.6 Herbage 22.8 18.6 13.2 13.5 21.9 18.3 13.7 13.3 Knapweed 71.2 76.9 83.2 82.4 71.8 77.2 82.4 83.1 Source: Popova (1960) Popova (67) found that deep plowing (Table XIX) brought about the elimina-tion of knapweed with subsequent vigorous grass growth. 52 TABLE XIX Effect of plowing (18 cm.) on C. diffusa i n the fruiting stage (% cover) Control 1955 1956 1957 1958 Just before Treatment 1955 Treatment 1956 1957 1958 Grasses Legumes Herbage Knapweed 1.0 - 0 14,. 0 85.0 7.4 0 10:. 2 82.4 8.5 0.6 10 .:4 80.5 6.'3 0.2 ••.'.9.1 84.4 1.3 0 1375' 85.2 78.4 0 20.2 1.4 80.6 0.7 17.4: 1.3 83.3 3.4 12,.7> 0.6 Source: Popova (1960) Popova (67) indicated that mowing actually increased the population of C_. diffusa .(Table XX) i n subsequent years following treatment. I t has been observed that the Centaurea species -are capable of resisting the action of mowing, not only by their rosette habit, but also by secondary flowering below.the original cutting height. Marsden-Jones and T u r r i l l (55) observed a period of secondary flowering i n the British knapweeds after mowing with capitula being produced i n September. TABLE XX. Mowing of C. diffusa i n the bolted stage (% cover) Control 1955 1956 . 1957 1958 Just before Treatment 1955 Treatment 1956 1957 1958 Grasses 20.6 22.2 18.2 15.3 22.4 Herbage . 32.0 . 25.5 23.4 22.3 30.2 Knapweed 47.4 52.3 58.4 62.4 47.4 9.1 3.1 2.5 18.5 13.5 11.4 72.4 83.4 86.1 Source: Popova (1960) 53 Popova (6-7) indicated that burning was an effective control measure for\" C. diffusa • (Table XXI).. He found that rosettes appeared after the-burn and developed the following year, but within two years the knapweed almost completely disappears from the grass sward due to the strong growth of the associated grass species. Zednai (105) found that a simulated range f i r e substantially reduced the germination percentage of .C. maculosa from 69% to .3%. TABLE XXI. Effect of burning on C. diffusa.in the fruiting stage (% cover) Control ^^f\"*\" ^ f \" ^ 0 ^ 6 Treatment Treatment 1956 1957 1958 1956 1957 1958 Grasses 12.5 10.4 8.3 ' 12.3 57.5 92.1 Herbage 10.4 10.1 9.2 9.6 . 5.3 6.3 Knapweed 77.1 79.5 82.5 78.1 37.2 1.6 Source: Popova (1960) r Hubbard (40,41) has established plots to determine the.effects of several variables on C. diffusa. These include picloram herbicide, com-'' plete protection from domestic grazing, and reseeding with crested wheat-grass. His results (Table XXII) indicate that forage production can be substantially increased through proper management practises i n dryland' situations. Where native grasses, have been substantially reduced, reseeding native range to crested wheatgrass shows considerable promise in increased forage yields. 54 TABLE XXII. • Effect of picloram, reseeding, and complete protection from grazing by domestic animals on the control of C_. diffusa Forage yield Treatment ( # ^ matter/acre) 1968 1969 1970 ' 1971 1. Check 77 221 234 252 .2. Seeded to crested wheatgrass 66 384 377 352 3. 4 oz. Picloram + crested wheatgrass 102 340 . 399 460 4. 8 oz. Picloram + crested wheatgrass 61 127 192 350 Source: Hubbard (1971) The Centaurea spp. are deep rooted biennials to short l i v e perennials which do not readily yield to the application of chemicals. However, Popova (67) found that 2,4-D at the rate of one kg per hectare completely eliminated diffuse knapweed and associated forbs i n badly infested pastures i n the Crimean (Table XXIII). Similarily, Furrer and Fertig (30) reported complete control of C. maculosa with 2,4-D amine or 2,4-D low volatile ester at the rate- of 1.5 lb. per acre and also with dicamba.-at the rate of 1 lb. per acre. . TABLE XXIII. 2,4-D (1 kg/hectare) applied to pasture infested with diffuse knapweed i n the rosette stage (% cover) n 4-. -, Just before „ , , . Control rn . . Treatment Treatment 12/6/56 14/6/57 . 16/6/58 12/6/56 21/6/56 14/6/57 16/6/58 Grasses 10.2% 16.9% 4.2% 10.2% 93.6% 91.2% 90.5% Herbage 19.4 .18.3 19.6 19.4 6.4 8.8 9.5 Knapweed 70.4 64.8' 76.5 70.4 0.0 0.0 0.0 Source: Popova (1960) 55 Renney and Hughes (72) indicated that t r i a l s using 2,4-D were begun i n 1952 in British Columbia to control diffuse knapweed. Control of C. diffusa was'temporary and the chemical application of 2,4-D did not prevent heavy seedling establishment i n the f a l l . Due to the re-infestation of C_. diffusa and C. maculosa after 2,4-D treatment i n British Columbia, further chemical t r i a l s were conducted. Renney and Hughes found that picloram (Tordon) exhibited the superior selective control of C. diffusa and C. maculosa on semiarid rangeland sites i n the interior of British Columbia. Picloram i s presently recommended at the rate of 6-8 oz. active ingredient per acre for the control of C. diffusa and C. maculosa i n the southern interior of British Columbia. They indicated that at least two years residual control \"has placed i t i n the category of an 'economical control practice'\". Ryerson and Sonder (76) have also indicated the beneficial effects of picloram for the control of spotted knapweed. Picloram, 4-amino-3,5,6-trichloropicolinic acid, has a relatively low mammalian toxicity and i s rapidly excreted from the animal body. The chemical i s extremely phytotoxic to broadleaf plant species with grass species only moderately affected. I t acts as a growth regulator. Sharma et a l . (83) found that the leaves and roots of Canada thi s t l e (Cirsium arvense (L.) Scop.) readily absorbed picloram and that the leaves, but not the roots, retained a substantial portion of the picloram absorbed. Scifres et al..(81) observed that the concentration of picloram, applied at the rate of h lb/a with 2,4,5-T at \\ lb/a, was reduced by 93% in herbaceous broadleaf plants by 30 days after herbicide 56 application. . They also found that picloram, at the above rate, usually dissipates from the- s o i l within 90 days under warm, dry conditions of semi arid rangelands (82). They suggested that the data indicates a low•probability of - picloram 'transport i n runoff water unless a heavy r a i n f a l l occurred'immediately after herbicide application. Goring and Hamaker (32)- found movement of picloram by volatilization and runoff i n f i e l d conditions was negligible. They indicated that the estimated half-order constants, ¥^ , of picloram i n Canada was in the order of -.0.2. Therefore, at rates of 1 oz. and 2 lb/acre the time required for the herbicide to decompose to a level of 0.01 oz/acre would vary from 4.5 months to 4.6 years. Picloram has been found to effectively control many undesirable woody plant species and many herbaceous weeds-. However, l i t t l e concern has been directed towards the effect of picloram on desirable forage grass species. Scifres et a l . (82) found that grasses absorbed picloram from the s o i l without forage production being adversely affected. More recent studies by Scifres and Halifax (80) showed root production of seedlings of switchgrass (Panicum virgatum L.) and sideoats grama (Bouteloua curtipendula (Michx.) Torr.) was decreased when 1 or 2 ppm picloram were placed on the s o i l surface. They also observed chlorotic and epinastic effects to the topgrowth of the seed-lings of these two grass•species. Scifres and Halifax stated: \"... picloram applied just prior to or following range reseeding or applied to badly depleted rangeland could further complicate rate of seedling establishment\". 57 The role of natural enemies-parasites, predators, and pathogens in reducing the population of a plant, or animal species is known as biological control (3). The biological control of weeds has been reviewed by a number of authors (3,4-2,43,96,97,99,107) and a number of striking successes have been recorded notably with the Opuntia spp. in Australia and the control of Hypericum perforatum in California. Huffaker (42) points out that the goal of biological control is the reduction of a pest population to a non-injurious level, and not eradication. A potential biocontrol agent must not only be capable of destruction of its plant host, but also be limited by the availability of its host as a food supply or reproduction.site. Huffaker and Andres (44) listed the following qualities of a good biological agent: 1. ability to k i l l plants or prevent reproduction in some direct or indirect way, 2. high ability to disperse and locate host plants, 3. good adaptation to the weed host and the. environmental conditions over a maximum part of terrain infested by the weed, 4. reproductive capacity sufficient to overtake the increase of its host without too much delay when for any reason control is temporarily short-circuited. Biological control is by its nature very selective and is most useful where a single aggressive weed is troublesome (3). Most of the serious weeds in North America are alien and their aggressiveness is usually attributed, in part, to the absence.of their natural enemies. The decision to release exotic natural enemies to control an alien weed must be preceded by extensive -studies to determine the safety or host specificity of the biocontrol agent. Harris and ZwOlfer (38,113) 58 described the limitations of standard methods of determining host spec i f i -city of insects (starvation and negative-ovipo'sition' /tests), and have suggested the studies should be broadened to include the following: 1. study of the insect's biology, including host plant records with particular attention to adaptations l i k e l y to rest r i c t host' ranges, 2. review of the plants attacked by related insects, . 3. determination of the laboratory host range of the insect, 4. investigations of the chemical or physical basis of host-plant recognition, 5. starvation tests on economic plants to confirm the limits of the previously established host range, 6. establishment of the insect's potential effectiveness for weed control. Since 1961 ZwOlfer (106,108,109,111,112) has conducted an extensive survey i n Europe to obtain information on phytophagous insects attacking wild Cynareae (Compositae) with the eventual hope of introdu-cing phytophagous insects into Canada for the biocontrol of thistles and knapweeds. The following l i s t s pf promising insects for the control of Centaurea diffusa and C. maculosa were developed from Zwolfer's reports (Tables XXIV and XXV). Numerous other phytophagous insects have been observed to damage Centaurea diffusa.and C. maculosa in,the European studies, but because of damage to economic crops such as artichoke or safflower or because of climatic limitations, have been excluded from the above l i s t s • . 59 TABLE XXIV. Potential insect agents for the biological control of C. diffusa Insect Type of damage Urophora aff i n i s (Trypetidae) galls within receptacle of flower head Metzneria sp. \"Z\" (Gelechiidae) larvae feed within receptacle and achenes of flower head Chaetorellia sp. '-'.(Trypetidae) larvae destroy young florets, ovarioles, receptacle, and achenes Sphenoptera jugoslavica (Buprestidae) ~ larvae mine within root-stock of the rosette and within the stem of the mature plant Pseudeucosma caecimaculana (Tortricidae) larvae mine within the peri-pheral tissues of the root Arima marginata (Galerucinae) adult and larvae feed on leaves Sphaeroderma rubidum (Chrysomelidae) adults feed on leaves and shoots, larvae mine within the leaves nr. Aceria grandis (Eriophydae) g a l l mite transforming single florets into s t e r i l e leaves Pterolonche sp. (Gelechidae) larvae mine within the. root Cyphocleonus tigrinus (Curailionidae) larvae mine within the stem Source: Zwolfer (1965, 1969, 1970, 1971) •60 TABLE XXV. Potential insect agents for the biological control of C. maculosa Insect Urophora aff i n i s (Trypetidae) Metzneria paucipunctella (Tortricidae) Euxanthoides straminea (Phalonidae) Chaetorellia hexachaeta (Trypetidae) Type.of damage galls within receptacle of flower heads larvae feed within flower heads and destroys achenes larvae mine within stems, roots, and flower buds larvae mine within receptacle and destroys achenes Terellia virens (Trypetidae) larvae destroy the achenes Source:- ZwOlfer (1965, 1969, 1970, 1971) r~\" ZwOlfer (110) has conducted an extensive study on the host spec i f i -city and l i f e history of Urophora affinis Frfld. (Dipt., Trypetidae). U. affinis adults appear i n the f i e l d during May-June and even i n July. When bud development occurs on the host plant the males and females assemble on the panicles of the Centaurea species where mating occurs. Females deposit eggs two or three days after emergence and deposit about one hundred and twenty eggs. The closed flower buds are carefully probed before deposition. Eggs are deposited singly .or i n small groups on the . small, undeveloped tubular flowers into the tissue of the receptacle. After three or four days the f i r s t instar hatches and penetrates the ovariole of the undeveloped -tubular flower where i t mines and causes the latter to develop into a fusiform g a l l (Figure 120. Normally one to three galls per flower head are formed but there may be up to eight 61 Fig. 12. Fusiform g a l l in the receptacle of a C. maculosa flower head I • I. . , . . r — u j MM i rt 1111111II {11«11 r I i; i l l l l l ini imimil l l l l l Fig. 13. Sclerotium in a C. diffusa root 62 galls formed in each flower head. The third ins tar larvae remain i n dia-pause within the g a l l .until early summer then pupate after turning so their head i s oriented to the distal opening of the g a l l . Adults emerge i n two or three weeks. The insects overwinters i n the pupal stage. ZwSlfer stated that \"the effect of U. affinis upon i t s host plant consists i n the destruction of achenes and in the deformation of the receptacle of the capitulum which leads to a reduction of the production of viable seeds\". Table XXVI summarizes the factors and responses involved i n the host specificity of the oviposition of U. a f f i n i s . These studies have led ZwOlfer to•conclude \"the central-European populations of U. aff i n i s (which are associated with C. maculosa) as \"safe\" for introduction to Canada for the following reasons: (a) The available f i e l d records suggested and extended oviposition tests made in the laboratory showed that \"the central-European population of U. aff i n i s are highly specialized i n their host selection; (b) An experimental analysis of the orientation of U. af f i n i s females showed that the females select a host plant and oviposition site using a com-bination of mainly or exclusively physical recognition tokens...; (c) Our study of U. affinis and other cecidocole Urophora spp. showed that in this genus the length of the ovipositor i s closely adapted to the dimensions of the oviposition s i t e . . . . . Thus, the heads of the•culti-vated Cynareae spp., artichoke and safflower are for mechanical reasons not accessible to U. affinis (Table XXVII); (d)...the cecidocole habits of the U. af f i n i s larvae are a criterion for the narrow host range of the species, since g a l l formation involves close physiological adapta-tions of the g a l l maker to the host.plant: (e) Inspite of the occurrence 63 TABLE XXVI. , Factors and responses involved i n the host specificity of op-position i n U_. affinis Phase of orientation - Property of the host plant Responses of the • which attracts or guides the U. aff i n i s female U. af f i n i s female Finding of host plant Visual aspect of stem, branches and stalks of the panicle. (Effect of olfactory stimuli not checked). Flying adults of U. a f f i n i s are attracted and land on the panicle Finding of ovi-position substrate Spatial arrangement and direction of the branches, especially•the angle formed by the branches and the vertical line. . (Effect of \"guiding 'rail\") Upward directed movement leads U. a f f i n i s to flower-buds External examination of oviposition sub-strate -Size and shape of closed ,flowerbuds (visual aspect and mechanically-tactile proper-ties Of secondary importance: Apical ruggedness of bud brought about by structure of external bracts. Visual pattern formed by external bracts (Of minor or no importance: chemical properties of the surface of the bud) Walking over the flower-bud which may'. be touched with proboscis or point of ovipositor sheath Attempts at probing (i.e. at intro-ducing ovipositor into the substrate Internal examination of oviposition sub-strate Thickness and hardness of the cover of the flower-bud (i.e. of middle and upper bracts) Internal structure- of the cavity of the flower-bud ^ Internal dimensions, probably mainly size of the undeveloped tubular florets Perforation of the cover of the flower bud i f the mechanical resistance of the cover can be overcome. Probing by means of the apical part of ovipositor. Deposition of eggs Source: Zwolfer (1970) TABLE XXVII. Length of ovipositor, and dimensions of oviposition site of some Urophora sp. Urophora sp. Host plant Length of the closed head of the host Length of the ovipositor attacking host U. a f f i n i s J r f l d Centaurea, diffusa, C. maculosa~ 4 (3)- 8 mm 4.5- -5.8 mm (1.7 mm)\"-U. siruna-seva Hg. C. s o l s t i t i a l i s 4-12 mm 5.0- 6.5 mm (1.9 mm) U. sp. nr s o l s t i t i a l i s L. (origin southern France) Carthamus lanatus (wild safflower) 18-25 mm 10.0-12.5 mm (3.8 mm) Urophora, undescribed sp. from southern Italy Cynara cardunculus (wild form) 35-45 mm 12.5-18.0 mm (4.9 mm) ^Values i n parentheses: Average length of ovipositor sheath for 10 individual females Source: ZwOlfer (1970) 65 of different ecotypes within the species of U. affinis i t can be assumed that the host pattern of the population associated with C. maculosa i s sufficiently stable to warrant introduction to North America\". Urophora a f f i n i s adults were released i n the Kamloops region i n the summer of 1970 under the direction of Dr. Peter Harris and with the co-operation of William Hubbard. U. affinis populations were released i n Centaurea maculosa and i n C. diffusa plant communities. Hubbard's (41) preliminary results indicated that approximately 3% of the C. maculosa,heads were attacked, but only 0.1% of the C. diffusa heads were attacked (Table XXVIII). The reason for the relatively low proportion of C. diffusa heads attacked may be due to the smaller flower buds of C. diffusa as compared to those of C_. maculosa. . Sweeps of the two release sites i n the spring of 1971 indicated that the f l y was capable of surviving the winter conditions i n the Kamloops area. Additional releases of U. affinis were conducted in'the spring and early summer of 1971. Dr. Peter Harris (36) has-determined that the U. affinis populations on C. diffusa and. C_. maculosa are increasing at the rate of approximately ten fold per year during the f i r s t two years since their release i n the Kamloops region. Another promising insect for the biological control of Centaurea spp. in British Columbia, Metzneria paucipunctella Zel., has been extensively studied by Englert.(21). Englert suggests that \" i t i s the most effective of the insects associated with the flowers of C. stoebe, a single larva being capable of destroying 95% of a l l viable achenes\". M. paucipunctella larvae do compete with U. affinis larvae and w i l l k i l l and feed on them if\"they come i n contact with the. U. af f i n i s larvae. 66 TABLE XXVIII. Urophora affinis,attack of Centaurea at B.C. release sites 2 (M m plots sampled i n the four cardinal directions) Distance from Average % Attack release point of heads Ipaces; . F a l l , 1971. Chase 0 3.80 (C. maculosa) 10 0.20 20 0.00 50 0.07 Walachin 0 4.70 (C. maculosa) 10 0.70 20 0.13 50 0.10 Pritchard 0 0.46 (C. diffusa).. 10 1.69 20 0.00 50 0.24 F a l l , 1970. Walachin 0 2.72 (C. maculosa) 5 1.94 10 2.42 20 0.32 Pritchard 0 0.36 (C. diffusa) 5 0.00 10 0.00 20 0.06 Source: Harris.(1971) Populations of M. paucipunctella w i l l be released i n the summer of 1972 at a site somewhat isolated from the Urophora release sites i n the Kamloops region. The use of plant pathogens as biological weed control agents has been recently reviewed by Wilson (96). The role.of plant pathogens in reducing plant populations i s well understood, but plant pathologists, have devoted l i t t l e attention to diseases of weeds. Wilson indicated the need for co-operation between workers i n biological control, usually entomologists, and plant pathologists. The potential damage of plant pathogens has been demonstrated by such devastating diseases as the Dutch elm disease and the late blight of potato. Recently Inman (47) stated: \"There i s a distinct possibility that certain selected plant pathogens may have sufficient pathogenicity on specific weeds to be of value as biological agents\". He has been studying the potential of a macrocyclic heteroecious rust., Lromyces rumicis (Schum) Wint.., as a biocontrol agent of Rumex species. ]hman suggests that U. rumicis has shown sufficient potential to warrant introduction into North America for further t r i a l s as a biological agent for curly dock. Dr. Peter Harris (35) has obtained a culture of Puccinia centaureae, an autoecious rust, from eastern Europe to study as a potential biocontrol agent for the Centaurea species i n Canada. However, the culture was not viable on arrival i n Canada.. I t i s hoped that addi-tional material w i l l be available shortly. 68 Guyot (34) indicated that Puccinia jaceae i s recorded on C. diffusa i n Bulgaria, Romania, Ukraine and Crimea where infection i s very heavy. Savile (79) surveyed the Puccinia species attacking Cardueae i n Europe and established Puccinia jaceae var. diffusae nov. by u t i l i z i n g the above collections. Guyot also indicated that C_. maculosa i s . a host of P. centaureae-vallesiacae i n Bulgaria,- Romania and Russia and a host of P_. jaceae i n Switzerland. P_. centaureae var. centaureae was also recorded on Centaurea species closely related to C. diffusa. C. diffusa plants infected with P. jaceae var. jaceae exhibited a resistant reaction and thus careful collection of the variety diffusae .would be of utmost necessity i f this organism i s to be considered as a potential biological control agent of C. diffusa i n British Columbia. Savile (79) conducted a survey of the native and introduced autoecious Puccinia species i n North America. Introduced rusts included P. centaureae var. centaureae which was observed on C_. nigra at Indian Point, Nova Scotia, i n July of 1965 by D.B.O. Savile. The rust was located i n only one colony even though the weed was locally common (79). P. a c r o p t i l i was recorded on Centaurea repens near Midway, British Columbia. Savile, i n regard to biological control of 9_. repens, suggests \"It might be worth seeking the importation of other biotypes for establishing-a degree of biological control of the host. The mor-phological distinctness of the rust strongly indicates that i t i s s t r i c t l y host-limited\". Numerous authors (3,46,86) suggests that the ideal method of weed control i s the combination of cultural, mechanical, herbicidal, and biological control which w i l l maintain the environment as detrimental to the weed as possible. Ryerson and Sonder (76) u t i l i z e d a herbicide band treatment as a means of seedbed preparation when studying the fe a s i b i l i t y of seeding desirable grasses onto range densely infested with spotted knapweed. They found that the establishment of thick-spike wheatgrass (Agropyron dasystachyurn) i n a l l herbicide banded strips was rated poor. They suggested that the poor establishment of the seeded wheatgrass was primarily due to competition from existing perennial grasses.. Where perennial grasses were absent,, establishment of the wheatgrass was good to excellent. They did not suggest that herbicide damage might have resulted i n poor establishment of the wheatgrass. Recent studies by Scifres and Halifax (80) have shown that picloram adversely affects the rate of grass seedling establishment when applied to reseeded range or range i n poor condition. Hubbard's (40) study also reveals that picloram retards grass establishment. Hubbard's integrated approach demonstrates, the benefits of combining herbicide treatment, reseeding, and protection from grazing on the control of C. diffusa i n the dry interior of British Columbia. 70 B. Experimental Methods A s o i l disturbance study was conducted to determine the effect of simulated cultivation on the population of C. diffusa. Two rangeland sites were chosen near Kamloops, B.C. The treatments were: one disturbance in the spring, two disturbances (spring and summer) and control. One meter square plots were set up in a randomized complete block design with three replications. The treatments were carried out with a shovel to simulate cultivation to a depth of approximately 10 cm. Near the end of the summer the number of plants anchored to the s o i l were recorded and seed was collected from plants which produced seed in each of the plots. Percent germination of the seed collected was determined by placing the seed (25 per dish) on moistened, sterilized-No. 1 Whatman f i l t e r paper i n glass petri dishes. The dishes were placed in a Seedburg germinator at 25°C. Data were analyzed by the analysis of variance. Studies were conducted to determine the effect of mowing on C. diffusa and C. maculosa. Rangeland sites at Summerland and Vernon, and Mara and Salmon River, were chosen for C. diffusa and C. maculosa respectively. Four treatments: mowing at bud stage, mowing at flowering, two mowings (bud and flower-stages) and a control were established in a randomized complete block design with three replications of each treatment at each of the two sites for each species. The meter square quadrats were mowed by hand with a small scythe to a height of 2 cm. Near the end of August the plots were observed and the number of seed producers s t i l l anchored to the s o i l were recorded. 71 Seed was collected from a l l plants which had produced seeds. Percent germination of the seed collected was determined by placing the seed (25 per dish) on moistened, sterilized No. 1 Whatman f i l t e r paper.in glass petri dishes. Data were analyzed by the analysis of variance. Long range studies were set up i n 1971 to determine the level of seed destruction required to reduce C. diffusa and C. maculosa to economic levels of infestation. The experiment was set up in-a randomized complete block design with three replications of ten treatments. The treatments were control, 10, 25, 40, 50, 60, 70, .80, 90 and 100% removal of seeds produced. No results have been obtained as data collection will.begin i n 1972. . Herbicide t r i a l s established i n 1966 and 1967 on C. diffusa and £• maculosa were scored in 1968, 1970 and 1971. The cover of knapweed on the plots was rated from 0 to 5 and the average of the replications of each treatment was calculated. C. Results and Discussion TABLE XXIX. Effects of s o i l disturbance on C_. diffusa populations Number of Plants/m „ . Treatment = = • : : Percent Mature Rosettes Seedlings Total Germ. 1. one dist. 3.00a5- 10.67a 4.67 18.33 97.17a 2. two dist. 0.00a 0.00 0.00a 0.00 0.0.0 3. • control 16.17 14.17a 1.17a 31.50 97.83a '''means i n the same column sharing the same letter do not d i f f e r s i g n i f i -cantly at the 5 % level according to Duncan's New Multiple Range Test. The data shown i n Table XXIX indicate? that .two relatively shallow cultivations, one i n early May and one i n early August, w i l l control . the population of C_. diffusa and prevent seed production. A single, relatively shallow cultivation, i n the early spring does significantly reduce the number of plants which reach maturity. However, the one cultivation has no effect on the number of rosettes, indicating the a b i l i t y of the rosettes to regenerate and continue growth. One shallow cultivation apparently brings viable seeds to the surface which germinate and produce a significant increase i n the number of seedlings. C. diffusa and C. maculosa are not commonly found i n irrigated pastures or irrigated hay fields. I t i s possible, therefore, that any range areas infested with knapweeds might be effectively controlled i f irrigation could be applied to these areas. Most rangelands produce 73 marginal returns however, and expensive irrigation systems are not commonly employed over vast acres of dryland range. The application.of f e r t i l i z e r to dense infestations of knapweed would have only adverse effects on potential control of the weed. Fe r t i l i z e r enhances the growth of Centaurea species (Table XXX). TABLE XXX. Shallow plowing with later f e r t i l i z e r application (10T fresh horse manure/hectare) (% cover) Control Just before Treatment Treatment 1955 1956 1957 1958 1955 1956 1957 1958 Grasses 14. 2' 15.1 17.1 6.5 21.5 10.6 10.1 10.6 Legume 0 0.3 0.1 0 0 1.2 0.-9 0.1 Herbage 29.4 28.3 32.3 28.9 24.2 18.5 19.3 6.4 Knapweed 56.4 56.3 50.5 64.6 54.3 69.7 69.7 82.9 Source: Popova (1960) However, rangeland i n good condition responds favourably to f e r t i l i -zers and their application i s important i n maintaining the competitive a b i l i t y of the grasses and other desirable species. Mowing i s a.management practise commonly used to reduce the seed production of weeds. The rosette habit of Centaurea species provides a means by which' i t can escape the effects of mowing, but this does not apply to the bolting and fruiting, stages. The results of the mowing t r i a l s (Table XXXI) indicate a s i g n i f i -cant reduction i n the number of plants which produced seeds i n a l l 74 TABLE XXXI. The effects of mowing on knapweed m. , . ' Number of 0 n . • Treatments , , • % Germination seed producers 1 mowing (bud stage) 7. 83a* 48. 00b 1 mowing (flower stage) 0. 25a 17. 92a 2 mowings (bud and flower stages,) 1. 17a 19. 92a Control 34. 33 91. 08b \"means in. the same column sharing the same l e t t e r do not di f f e r significantly at the 5% level according to Duncan's New Multiple Range Test. treatments and the percent germination was significantly reduced by treatments two and three when compared to the control. There was no significant difference between the two species i n these studies. Therefore Table XXXI consists of the combined means of both Centaurea species. C. diffusa seed which was obtained from an area which had been burned by a forest f i r e in mid August of 1971 was not viable. These observations suggest that burning could be a useful control measure against the Centaurea species i n British Columbia. However, due to the disadvantages associated with burning such as the d i f f i c u l t y i n achieving a uniform burn and the control of the f i r e , i t s potential use as a means of controlling knapweed infestations i s limited. Good management practises for native range are not costly procedures and are often not thought of as an annual expense by many 75. ranchers. Young (102) has suggested a cost of $12.00 to $15.00 per acre for the complete development of a management unit for range improvement including: plowing, seeding, water development, fencing, non use and interest on investment. . However, these costs are generally spread over a number of years and consequently are not a heavy burden to the rancher. The major limiting factor i n the use of cultural control methods for rangeland i s the d i f f i c u l t y with rough terrain. Much of the area cannot be traversed by common farm implements. Through a co-operative effort of the British Columbia Departments of Agriculture, Forestry, and Highways and the Department of Plant Science, University of British Columbia, a knapweed control program was conducted i n 1970 and 1971 i n the East Kootenay region of British Columbia. The herbicide, picloram, applied to infestations of knapweed i n 1970 gave excellent results with minimum need for re-treatment of 'fringe' areas in 1971. Table XXXII illustrates the acres infested with Centaurea species which were treated with picloram i n the East Kootenay region of British Columbia. TABLE XXXII. Results of the East Kootenay Knapweed Control Control Program 1970-1971 Species Acres treated Rate CLb ai/a) 1970. 1971* Total C. repens C. diffusa and C. maculosa:, C. nigra 1.5 0.5 0.5 27 4 194 58.8 100 sq. f t . . 31 252.8 100 sq. f t . * acreage does not include small ' fringe'. areas around plots treated i n 1970 which were retreated i n 1971. 76 It i s hoped that this program has controlled the Centaurea species i n the East Kootenay region and w i l l prevent vast acreages of rangeland from becoming infested with these troublesome aggressive knapweed.species. The Department of highways i n the Kamloops, the Cache Creek and the Okanagan' Valley regions have begun to use picloram i n their road-side spraying programs to control such troublesome weeds as leafy spurge and the knapweed species along road rights-of-way. A program to control C. diffusa on the Research Station at Summerland was i n i t i a - : ted i n 1971.. Similarly, cseveralvo.r concerned ranchers have also-init i a t e d spraying programs with picloram to attempt,to control the • spread of knapweed infestation. The rather high cost of picloram, $53.50 per gallon with 2.4 lb. active ingredient or approximately $10.70 -per acre, has greatly limited i t s potential use i n controlling the Centaurea species i n the interior of British Columbia-. The returns per. acre would need to be sufficiently high before applications of this expensive chemical over extensive areas of semi-arid rangeland were initiated. The returns per acre of dry-land' ranges are marginal and do not generally warrant application of picloram to control the knapweed species. Costs can be substantially reduced with the use of airplanes when large acreages are treated. However, the application of such a phytoxic chemical over vast areas of semi-arid rangeland could cause considerable damage. Undoubtedly, numerous forbs,..-; susceptible to picloram, would become endangered species I f broad appli-cation of the herbicide was employed. The relatively long persistence of picloram i n the s o i l and i t s phytoxic effects on such forage crops 77 as a l f a l f a has further limited i t s use in areas which may be placed i n forage production i n the near future. The herbicide, picloram, does provide encouragement in the control of the knapweed species in the interior of British Columbia, especially along roadsides and scattered infestations. . However, due to i t s high costs and possible adverse ecological effects, the major infestations cannot be economically controlled using herbicides. The perimeter areas of the knapweed infestations may be effectively controlled and the spread of knapweed infestations can be restrained with the effective use of herbicides. In the late spring of 1971rthe author observed a number of C. diffusa plants i n the bolted stage which appeared wilted and eventually died before seed production. The affected plants were located in a rangeland infestation east of Vernon, British Columbia. Close examination of the affected plants indicated considerable swelling at the crown region of the tap root. Dissection of the roots of affected plants revealed a black sclerotium-like body (Figure 13). Numerous other sclerotia were observed on and just beneath the root epidermis of affected plants. Two fungal organisms were consistently isolated from the diseased plant material. Before identification of the fungus which produced the sclerotia could be attempted the dormancy of the sclerotia had to be broken. The breaking of the sclerotia dormancy and subsequent production of apothecia was accomplished by placing the sclerotia, growing on sterilized potato dextrose agar, into cold treatments of 4°C, 0°C or -24°C for periods of one or two months. The treated sclerotia were then placed in sterilized 78 moist greenhouse, s o i l . The Koch-postulates as described by Barnes (7) were carried out with the two fungal i s o l a t e s ; Table. XXXIII\" indicates that apothecia were formed from s c l e r o t i a receiving the 4°C and 0°C treatments but not from those receiving the -24°C treatments (Figure 14). Cultures of the s c l e r o t i a and apothecia with ascospores were i d e n t i f i e d by M.\"E. E l l i o t t as S c l e r o t i n i a sclerotiorum (Lib.) de Bary (20) (Figure 15). iiiiniiuiiiiiiim 0NV19M3 Ni 3uvi^ Fig. 14. Apothecia produced from cold treated sclerotia Fig. 15. Plate culture of S_. sclerotiorium 80 TABLE XXXIII. . Treatments u t i l i z e d to break sclerotia dormancy Temperature Treatment Duration Min. time for apothecia to appear after being placed in the s o i l >2 months 1 month 1 month 1 month 2 months 2 months none formed none formed 98 days 46 days 33 days 68 days The fungal organism .which was concomitantly isolated from the diseased plant material with S_. sclerotiorum has been identified as a new species, Microsphaeropsis centaurea^Morgan-Jones sp. nov. (in.ed.) (68) (Figure 16). Table XXIV illustrates the results of following Koch's postulates with S_. sclerotiorum and M. centaureaeon C. diffusa and C. maculosa. One possible means of natural spread of the disease organisms may be due to a root coccid, Phenacoccus sp. nr. solani Ferris, which were collected from the roots of a number of C_. diffusa and C. maculosa plants i n the dry interior of British Columbia. The root coccid was identified by W.R. Richards (73). The w i l t fungus, Sclerotinia sclerotiorum, i s widely distributed and causes considerable damage to vegetable crops such as beans, lettuce and celery. Conners (14) l i s t s S_. sclerotiorum as pathogenic on such Fig. 16. Plate culture of M. centaureae TABLE XXXIV. Determination of pathogenicity of Sclerotinia sclerotiorum and Microsphaeropsis centaureae on Centaurea diffusa.and C. maculosa Fungus Organism Host material Method of inoculation Disease Symptoms Produced Re-isolation S_. sclerotiorum Mature C. maculosa Rosette C. maculosa Mature C. diffusa . Rosette C. diffusa s t e r i l e hyphae suspension to s o i l plus wounding of host root material wilted wilted sclerotia i n root + + S. sclerotiorum Mature C. maculosa Rosette C. maculosa Mature C. diffusa Rosette C. diffusa s t e r i l e hyphae suspension to s o i l wilted wilted sclerotia i n root S_. sclerotiorum. plus M. centaureae Mature C. maculosa Rosette C. diffusa Mature C. maculosa Rosette C. diffusa st e r i l e hyphae suspensions of both organisms to s o i l S. sclerotiorum Rosette C. maculosa Seedling C. maculosa Rosette C. diffusa Seedling C. diffusa st e r i l e hyphae suspension to leaf surface wilted (death) wilted (death) wilted (death) wilted (death) + + + + M. centaureae Rosette C_. maculosa st e r i l e spore suspension Seedling C. maculosa sprayed onto leaf surfaces Rosette C. diffusa Seedling C. diffusa \"leaf spotting followed' by death/.'-\"leaf spotting followed by death.\";-Leaf spotting symptoms were repeated in a replication of the i n i t i a l experiment. CO 83 valuable crops as a l f a l f a (Medicago sativa L.) and sunflower (Helianthus annus L.). The ubiquitous nature of this plant pathogen and i t s broad host range decisively l i m i t i t s potential as a biological control agent of the Centaurea species i n the interior of British Columbia. The'United States Department of Agriculture (2) has l i s t e d plant diseases which have been observed on species of the genus Centaurea. Table XXXV l i s t s those diseases observed on Centaurea species other than C. americana, C_. cyanus, C. cineraria and C_. montana. TABLE XXXV. Plant diseases observed on Centaurea species i n North America Organism State Disease Albugo tragopoyonis . Tex. white-rust Erysiphe cichorearum Calif. powdery mildew Meloidogyne sp. Ohio root-knot nematode Phymatotrichum omnivorum .. Tex. root rot Plasmopara halstedii - Iowa . downy mildew Puccinia cyani Md. and Tex. rust Rhizoctonia solani Tex. root and stem rot Sclerotinia sclerotiorum Tex. stem rot, w i l t Sclerotium r o l f s i i N.J. southern blight Yellows (Chlorogenus callistephi) Pa virus Source: U.S.D.A. Handbook 165 (I960) Dr. C.G. Shaw \"(84) has indicated that there are no additional records of parasitic or pathogenic fungi on species of the genus • 84 Centaurea. Conners (14) did not l i s t any plant diseases of C. diffusa or C. maculosa i n Canada. The report of S_. sclerotiorum on Centaurea in Texas would not have been observed on C_. diffusa nor C_. maculosa as these species have not been recorded in that state; ' Microsphaeropsis centaureae i s pathogenic to juvenile leaves of C. diffusa and C. maculosa. The fungus produces necrotic lesions and eventual leaf death of infected plants (Figures 17 and 18). Sutton (90) reports Microsphaeropsis. callestra (H. Syd) nov. comb, as parasitic on leaves of Eucalyptus haemastoma:.in Australia. He describes the lesions as follows: \"...up to 0.5 cm. diam. , mostly circular but occasionally irregular, sometimes confluent but more frequently separate,- appearance similar on both sides pf the leaf, edge raised, sharply delimited from the healthy tissue by a brown to purplish brown l i n e , surrounded by a diffuse halo of brown to purplish brown discoloration, central lesion tissue cream throughout or brownish towards the periphery\". The leaf spots observed on C_. diffusa and C. maculosa infected with M. centaureae. appear very similar to those described by Sutton. Campbell (11) has reported sclerotia of S_. sclerotiorum as being parasitized by Coniothyrium minitans and indicated the possibilities of biological control of Sclerotinia species by Coniothyrium minitans. The genus Microsphaeropsis Holm i s based on Coniothyrium olivaceum Bon. (61). This•suggests that M. centaureae may be.parasitic on the sclerotia of S. sclerotiorum. However, i t does not explain why both fungal organisms were consistently isolated from a l l portions of the diseased host material. Both fungus organisms Care, parasitic .tcTC. 7diffusa \"and C. maculosa. Fig. 17. Leaf spot on C. diffusa Fig. 18. Leaf spot on C_. maculosa 86 The association between these two fungus species requires further study. Also the potential of M. centaureae as a biological agent of Centaurea species i n the interior of British Columbia needs to be explored more f u l l y . The economics of biological control have been reviewed by a number of authors (5,46,86). They have indicated that the annual financial benefit i s exceedingly high with costs limited to exploration, research, and introduction of the biological agent. Once control i s achieved, no expense or very l i t t l e recurrent annual expenses are incurred. Hussey (46) points out \"that biological control...is particularly useful and suitable for pest control i n situations where costly chemical controls are out of the question\". There has been, i n the past, a lack of accurate economic information which has lead to considerable criticism of biologi-cal control. Therefore, accurate balance sheets of biocontrol attempts should be maintained to i l l u s t r a t e the benefits of biological control. The data in Table XXXVI illustrates the order of magnitude of the costs of biological control of a weed and include the approximate number of scientist-years spent on i t plus the cost of overseas exploration. Harris (37) suggested that: \"In round figures the cost of biological control of a weed may be as much as $500,000 i f i t i s possible to use agents that have been shown safe and effective elsewhere and twice this sum i f i t i s necessary to do the development work\". These costs compare very favourably to those for developing a new herbicide but,as Harris points out, biological control does not have the continued costs of expense and labor once the biocontrol agent becomes self-regulating. 87 TABLE XXXVI. Evaluation of biological control measures against weeds i n Canada Weed Start of Prog, in Canada Estimated costs5' Abroad Canada Degree of Success5'\"'5 Carduus?;.s\"p. Cirsium. arvense Euphorbia sp. Hypericum perforatum Linaria vulgaris Senecio jacobaea. 1968 1963 1965 1952 1957 1961 45 20 tt 45 320 t 25 100 - t 10 410 (threat) tttt (forage) ttt 10 120 t 10 .220 (forage) ttt (cattle) tttt * thousands of dollars ** - No control t Slight pest reduction or too early for evaluation. tt Local control, distribution restricted or not f u l l y investigated. ttt Control widespread but local damage occurs tttt Control complete Source: Harris (1971) Huffaker and Andres (44) state: \"where effective, biological control by introduced natural enemies i s cheap, permanent i n nature, without need for recurrent expense, and does not add toxic pollutants to the environment, endangering non-target organisms, plant or animal\". The Centaurea species are common along roads and readily invade misused dryland ranges i n the southern interior of British Columbia. Each method of control has i t s limitations. The use of cultural control such as mowing or cultivation i s generally not applicable for knapweed control. Herbicides are'useful management tools , but the high costs of picloram have limited i t s use over large areas of range-land.. Biological control of the Centaurea species has just recently been in i t i a t e d and i t w i l l be 6 to 10 years before the returns w i l l be appreciated i f the biological control agents become sufficiently well established. Control, by whatever means, does not eliminate the possibility of a 'resistant' weed replacing the knapweed i n the dryland• range environment. Control methods must be accompanied with appropriate management.practices i n order to maintain vigorous stands of useful forage. Proper range management includes moderate grazing of the range to maintain vigorous grass growth and to protect the plant crowns.' Heavy grazing should always be followed by a sufficient rest period to allow depleted food reserves to be b u i l t up i n the roots of the forage species. Even.distribution of grazing animals i s also essential on well managed range (51,57). 89 Fig. 19. Re-establishment of C. diffusa after herbicide treatment at OK Falls, B.C. 5 4_ Weed 3-Density Rating 2-(0-5) 1-0_ iv / V • * y / y r 1967 —I 1 1968 1969 Time (yr.) 1970 1971 f time of herbicide application A 2,4-D amine @ 2 lb/a B picloram @ 8 oz/a Fig. 20. Re-establishment of C. maculosa after herbicide treatment at Chase, B.C. 5 4 -Weed 3 -Density Rating 2 _ (0-5) 1 -0 . B 1967 T T 1968 1969 Time (yr.) —r— 1970 —J\" 1971 ^ time of herbicide application A 2,4-D amine @ 2 lb/a B picloram @ 8 oz/a 90 Figures' 19 and 20 i l l u s t r a t e the residual effect of picloram i n controlling knapweeds as compared with 2,4—0 amine. The dry OK Falls site i s i n extremely poor condition. The apparent poor response to the herbicide treatment and relatively rapid re-establishment of C. diffusa i s partially due to the lack of grass establishment after the.herbicide treatments.' The Chase site was i n better condition prior to herbicide treatment and native grasses were capable of increased production as the competition of C. maculosa was greatly reduced i n the picloram treated plots. Grazing was not controlled at either site and as a result of subsequent overgrazing of the treated areas the knapweed species were capable of rapid re-establishment. These studies indicate the need for good management techniques to follow-herbicide treatment of knapweed infested areas. It can be recommended that control of C. diffusa and C.. maculosa , can be achieved by cultural- methods such as cultivation and irrigation on only those areas that are amenable to extensive cultivation. Herbici-dal control can be u t i l i z e d to control roadside infestation and to prevent further spread of the knapweed from the perimeter of the existing infestation. The f e a s i b i l i t y of biological control i s encouraging and the development of phytophagous and pathogenic agents could economically reduce the knapweed infestation i n the interior of British Columbia. 91 V. BIBLIOGRAPHY . 1. Abrams, L. and R.S. Ferris. 1960. Illustrated f l o r a of the Pacific States. Vol. IV. Stanford University Press. Stanford, California, pp. 543-544. 2. Anonymous. 1960. Index of plant diseases i n the United States. U.S.D.A. Pub. No. 165. Washington, D.C. pp. 70. 3. Anonymous. 1968. Weed Control. National Academy of Sciences. Pub. No. 1597. Washington, D.C. 4. Anonymous. 1970. Selected Weeds of the United States. U.S.D.A. Agr. Handbook No. 366. Washington, D.C. pp. 380-383. 5. Anonymous. 1971. Biological Control. 41st Annual Report 1969-1970. Commonwealth Agricultural Bureaux, pp. 24-29. 6. Atkinson, T.G. and V.C. Brink. 1953. Progress report.on the biology and control of diffuse knapweed (Centaurea- diffusa Lam.) i n British Columbia. Department of Agronomy, University of British Columbia. 7. Barnes, E.H. 1968. Atlas and Manual of Plant Pathology. Appleton-Century-Crofts, New York. 8. Braun-Blanquet, J. 1951. Pflanzensoziologie: Grundzuge der Vegetationskunde. 2nd ed.-, Springer-Verlag, Vienna. 9. British Columbia Provincial Government. 1971. Herbarium. Provincial Museum, Victoria, B.C. 10. Britton, N.L. and H.C. Brown.. 1923. An illustrated f l o r a of the Northern United States, Canada, and the British Possessions. Vol. I I I . \"Lancaster Press Inc.' Lancaster, P.A. p. 558. 11. Campbell, W.A. 1947. • A new species of Coniothyrium parasitic on sclerotia. Mycologia 39: 190-195. 12. Canada Department of Agriculture. 1971. Phanerogamic Herbarium. Plant Research Institute, Ottawa, Ontario. 13. Cavallito, C.J. and J.H. Bailey. 1949. An antibacterial principle from Centaurea maculosa., J. Bacteriology 57: 207-212. •92 14. Conners, I.L. 1967. An annotated index of plant diseases i n Canada. Canada Department of Agriculture Pub. 1251. Queen's Printer, Ottawa. 15. Coxworth, E.C.M. J.M. B e l l , and R. Ashford. 1969. Preliminary evaluation of Russian t h i s t l e , kochia and garden atriplex as potential high protein content seed crops for semiarid areas. Can. J. Plant. Sci.. 49: 427-434. 16. Davis, R.J. 1952. Flora of Idaho. Wm. C. Brown Co. Dubuque, Iowa. pp. 713-715. 17. Deam, CC. 1940. Flora of Indiana. Wm. B. Burford Printing Co., Indianapolis, Ind. p. 1104. 18. Earle, F.R. and Q. Jones. 1962. Analysis of seed samples from 113 plant families. Econ. Bot. 16: 221-250. 19. Eastham, J.W. 1947. Supplement to 'Flora of British Columbia'. Special Pub. No. 1, B.C. Prov. Museum. Department of Education, Victoria, B.C. pp. 107-108. 20. E l l i o t t , M.E. 1972. Personal Communication.- Experimental Farm, Canada Dept. Agriculture, Ottawa, Ontario. • 21. Englert., W.' 1971. Metzneria paucipunctella Zel. (Gelechiidae, Lepidoptera): a potential insect for the biological control of Centaurea,stoebe L. i n Canada. Prog. Rep. Commonw. Inst. Bi o l . Control 28: 1-12. 22. Feldman, L.J.' and E.G. Cutter. 1970. Regulation of leaf form i n Centaurea s o l s t i t i a l i s L. I. Leaf development on whole plants i n ste r i l e culture. Bot. Gaz. 131: 31-39. 23. Feldman, L.J. and E.G. Cutter. 1970. Regulation of leaf form i n Centaurea s o l s t i t i a l i s L. I I . The development potentiali-ties of excised leaf primordia i n st e r i l e culture. Bot. Gaz. 131: 39-49. 24. Fernald, M.L. 1950. Gray's .Manual, of Botany. American Book Company, New York. pp. 1544-1545. 25. Fertig, S.N. 1968. Broadleaf weed problems of the Northeast Weed Control Conference. Proc. 22nd Northeast Weed Control Conf. pp. 19-26. 9'3' 26. Fletcher, R.A. 1961. A growth inhibitor found i n Centaurea spp.. M.Sc. Thesis, Dept. Plant Sci., University of British Columbia. 27. Fletcher, R.A. and A.J. Renney. 1963. A growth inhibitor found i n Centaurea spp.. Can. J. Plant Sci. .43: 475-481. 28. Frankton-, C. 1967. Weeds of Canada. Canada Dept. Agr. Pub. 948. Ottawa, Ontario, p. 170. 29. Frenkel, R.E. 1970. Ruderal Vegetation Along Some California Road-sides. University of Calif. Pub. i n Geography Vol. 20. Univ. Calif. Press.. 30. Furrer, A.H. and S.N. Fertig. 1965. Progress report on herbicide treatments for the control of spotted knapweed (Centaurea maculosa). Proc. 19th NEast Weed- Control Conf. pp. 324-326. 31. Gilkey, H.M. 1957. Weeds of the Pacific Northwest. Oregon State College, Corvallis, Oregon, p. 403. 32. Goring, C.A.I, and J.W. Hamaker. 1971. The degradation and movement of picloram i n s o i l and water. Down to Earth 27: 12-15. 33. Groh, H. 1943. Canadian Weed Survey. 2nd Annual Rep. Canada Dept. Agr. 34. Guyot, A.L. 1967. Les rouilles des Centauries. Uredineana 6: 59-161. 35. Harris, P. 1970. Personal communication. Acting Director, Research Station, Canada Dept. Agriculture,' B e l l e v i l l e , Ontario. 36. Harris, P. 1971. . Personal communication.-37. Harris, P. 1971. Current approaches to biological control of weeds. In Biological control programmes against insects and weeds i n Canada .1959-1968. Tech. Communication No. 4, Commonw. Inst. Biol. Control, C.A.B. Lamport Gilbert Printers Ltd., Reading, England, pp. .67-76. 38. Harris, P. and H. ZwOlfer. 1968. Screening of phytophagous insects for biological control of weeds. Can. Ent. 100: 295-303. 39. Howell, J.T. 1959.' Distributional data on weedy thistles i n western North America. Leaflets Western Botany 9: 17-29. 94 40. Hubbard, W.A. 1970. Knapweed Control. Canadex 641. Canada Dept. Agr. . 41. Hubbard, W.A. 1971. Biological Weed Control. Annual Rep. Research Station, Canada Dept. Agr., Kamloops, B.C. 42. Huffaker, C.B. 1962. Some concepts on the ecological basis of biological control of weeds. Can. Int. 94: 507-514. 43. Huffaker, C.B. 1964. Fundamentals of biological weed control. In Biological Control of Insect Pests and Weeds. P. Debach ' (ed.) Reinhold Pub. Corp. pp. 631-649. 44. Huffaker, C.B. and L.A. Andres. 197Q. Biological weed control using insects. FAO International Conf. on Weed Control. Davis, Calif, pp. 436-449. 45. Hultin, E. and K. Torsell. 1965. Alkaloid screening of Swedish plants.- Phytochemistry 4:. 425-4-33. 46. Hussey, N.W. 1970. Some economic considerations i n the future development of biological control. In Technological Economics of Crop Protection and Pest Control. J.G. Gregory, (ed.) S.C.I. Monograph No. 36, Soc. Chem. Indus-try, London, pp. 109-118. -47. Inman,, R.E. 1970. Observations on the biology of Rumex rust Uromyces- rumicis (Schum) Wint., Bot. Gaz. 131: 234-241. 48. Isely, D. 1960. Weed identification and control i n the North Central States. Iowa State University Press. Iowa., p. 172. 49. Jones, G.N. 1963. . Flora of I l l i n o i s . University of Notre Dame Press, Notre Dame, Indiana, pp. 269-270. 50. Lakela, 0.\" 1965. A Flora of Northeastern Minnesota. University of Minnesota Press. Minneapolis, Minnesota, pp. 398-511. 51. Lodge, R.W., J.B. Campbell, S. Smoliak, and A. Johnston. 1971. Management of the Western Range. Canada Dept. Agr. Pub. 1425.-52. Louveaux, J. 1971. Personal communication. Station de recherches apicoles, Bures-sur-Yvette, France. 53. Lovell, H.B. 1966.'. Let's talk about honey plants. Glean. Bee Cult. 94: 611-612. •' 1 95 54. Lovell, H.B. 1967. Let's talk about honey plants. Glean Bee Cult. 95:' 36-55. Mars den-Jones, E.M. and W.B.-Turrill.- 1954. British Knapweeds. Ray Soc. No. 138. Bernard Quaritch Ltd., London. 56. Maurizio, A. and I. Grafl. 1969. Das Trachtpflanzenbuch. Nektar und Pollen-die wichtigsten Nahrungsquellen die Honigbiene. Franz Ehrenwirth Verlag KG. Munchen, Germany pp. 162-164. 57. McLean, A. and L. Marchand. 1968. Grassland ranges i n the southern interior of British Columbia. Canada Dept. Agr. Pub. 1319. 58. Monya, M., M. Sabau, and G. Racz. 1968. Die antibiotiache wirkung auszugen aus Centaurea - arten. Planta Med. 16: 58-62. 59. Moore, R.J. 1969. How weedy thistles came ,to Canada. Greenhouse-Garden-Grass 8: 1-2. 60. Moore, R.J. and C. Frankton. 1954. Cytotaxonomy of three species of Centaurea adventive i n Canada. Can. J. Bot. 32: 182-186. 61. Morgan-Jones, G. 1972. Personal communication, Department of Biology, Waterloo University, Waterloo. 62. Muenscher, W.C. 1955. Weeds. The Macmillan Company, New York. pp. 441-442. 63. Muenscher, W.C. 1962. Poisonous Plants of the United States. The Macmillan Company, New York. pp. 13-20. 64. Ormrod, D.J. 1966. Surface anatomy of weed leaves with particular reference to stomata. M.Sc. Thesis, Dept. Plant Sci., University of British Columbia, Vancouver. 65. Ormrod, D.J. and A.J. Renney. 1968. A survey of weed leaf stomata and trichomes; Can. J: Plant Sci. 48:- 197-209. 66. Peck, M.E. 1961. A manual of the higher plants of Oregon. Binfords and Mort Pub. Portland, Oregon, pp. 846-848. 67. Popova, A.Ya. 1960. Centaurea diffusa Lam., a steppe-pasture weed-in the Crimea. (Russian) Bot. Zhur. (Moscow) 45: 1207-1213. 68. Popova, A.Ya. 1964. Composition of nutritional substances of Centaurea diffusa Lam. i n Crimea, (trans, from Russian) Bot. Zhur. .(Moscow) 49: 863-865. S.6. 69. Racz, G. , G.' Augustin, and E. Racz-Kotilla. 1963. Valorificarea-speciilor indigene de centaura ca plante medicinale cu continut i n princippii.amare. Lucrarile Grandini Bot Bucuresti 1961/1962 (1): 521-524. 70. Radford, A.E., H.E. Ahiles, and CR. Bell. 1964. Guide to the vascular flora of the Carolinas. Book Exchange, University of North Carolina, Chapel H i l l , N.C. p. 327.' 71. Renney, A.J. 1959'. Centaurea spp. infestations i n British Columbia. Proc. Joint Meeting N. \"Central Weed Control Conf. 16 and W. Canada Weed Control Conf. 10: 18-19. 72. Renney, A.J. and E-C. Hughes. 1969. Control of knapweed, Centaurea. species, i n British\" Columbia with Tordon herbicides. Down to Earth 24: 6-8. 73. Richards, W.R. 1971. Insect Identification .Report 71-509. Canada Dept. Agr. Research Branch, Entomology Research Institute. 74. Roche, B.F.Jr. 1967. Range and pasture weed survey. Forestry and Range Management, Washington State University, Pullman, Wash. 75. - Rydberg, P.A. 1932. Flora of the Prairies and Plains of Central North America. New York Botanical Garden, New York.-pp. '885-886. 76. Ryerson, D.E. and L.W. Sonder. 1967. The effects of herbicide control of spotted knapweed (Centaurea maculosa), and other forbs on forage production i n a grassland-park type i n western Montana. West. Weed Control Conf. Res. Prog. Rep. pp. 16-17. 77. Saarisalo-Taubert, A. 1966. A study of hybridization i n Centaurea, section Jacea,. i n eastern Fennoscandia. Ann. Bot. Fenn. 3: 86-95. 78. Savile, D.B.O. 1970. Some Eurasian Puccinia species attacking Cardueae. Can. J. Bot. 48: 1553-1566.' 79. Savile, D.B.O. 1970. Autoecious Puccinia species attacking Cardueae i n North America. Can. J. Bot. 48: 1567-1584. 80. Scifres, C.J. and J.C. Halifax. 1972. Root production of seedling grasses i n s o i l containing picloram. J. Range Management 25: 44-46. 97 81. Scifres, C.J., R.R. Hahn, and M.G. Merkle. 1971. Dissipation of picloram from vegetation of semiarid rangelands. Weed Sci. 19: 329-332. 82. Scifres, C.J., R.R. Hahn, J. Diaz-colon, and M.G. Merkle. 1971. Picloram persistence i n semiarid rangeland soils and water. Weed Sci. 19: 381-384. 83. Sharma, M.P., F.Y. Chang, and W.H. Vanden Born. 1971. Penetration and translocation of picloram i n Canada t h i s t l e . Weed Sci. 19: 349-355. 84. Shaw, C.G. 1971. Personal communication. Dept. Plant Pathology, Washington State University. 85. Simmonds, F.J. 1968. Biological control of pests of veterinary importance. Vet. Bull. 37: 71-85. 86. Simmonds, F.J. 1967. Economics of biological control. J. Roy. Soc. Arts 115: 880-898. 87. Smith, H.V. 1966. Michigan Wildflowers. Cranbrook Institute of Science, Bloomfield H i l l s , Michigan, p. 431. 88. Stevens, W.C. 1948. Kansas Wild Flowers. University of Kansas Press. Lawrence, Kansas, p. 428. 89-. Steyermark, J.A. 1963. Flora of Missouri. Iowa State University Press. Ames., Iowa. pp. 1628-1654. 90. Sutton, B.C. 1971. Coelomycetes. IV. The genus Harknessia, and similar fungi on Eucalyptus. . Commonw. Mycol. Inst.., Kew. Mycological Papers 123: 35-37. 91. Taylor, R.L. 1972. Personal communication. Dept. Plant Sci., Uni-, versify of British Columbia, Vancouver. 92. Timbers, G.E. and R.P. Hocking. 1971. A temperature gradient bar for seed germination and cold hardiness.. Can. J. Plant Sci. 51: 434-437. 93. University of British Columbia. 1971. Herbarium. Dept. Botany, Vancouver.-94. Vorobiov, M. Ye. 1960. Pro deyaki biolohichn osoblivoste dvorichnykh bur'yaniv Predunaeskoho stepu. Ukrain. Bot. Zhur. 17: 43-49. : 98 • 95. Washington State University. 1971. Herbarium. Dept. Botany, Pullman, Wash 96. Wilson, C.L. 1969. Use of plant pathogens i n weed control. Ann. Rev.' Phytopathology 7: 411-434. 97. Wilson, F. 1960. The future of biological control. Rep. Commonw. Entomol. Conf. 7th. London, pp. 72-79. 98. Wilson, F. 1964'. .The biological control of weeds. Ann. Rev. Ent. 9: 225-244. 99. Wilson, F. 1970. Biology to combat pestilence. New Scientist 8: 72-74, 100. Wodehouse^ R.P. 1945. Hayfever plants. Chronica Botanica Company. Waltham, Mass.. p. 162.' 101. Young, J.A.'. 1969. Areas and.species encountered i n herbaceous range weed control. Ann. Prog. Rep. Res. Committee West. Soc. Weed Sci.. pp. 12-14. 102. Young, J.A. 1969. Range improvement st a t i s t i c s . Ann. Prog. Rep. Res. Committee West. Soc. Weed Sci.. pp.- 14-16. 103. Young, S., W.W. Brown, and B. KLinger. 1970. Nigropallidal Encephalomalacia i n horses fed Russian knapweed (Centaurea. repens L.). Am. J. Vet. Res. 31: 1393-1404. 104. Young, S. , W.W. Brown, and B. KLinger. 1970. Nigropallidal Encephalomalacia i n horses caused by ingestion of weeds of the genus Centaurea J.A.V.M.A. 157: 1602-1605. 105. Zednai, J.G. 1968. Studies on the biology and control of . Centaurea maculosa Lam,. B.Sc. (Agri.) Thesis. Dept. Plant Sci., University of British Columbia, Vancouver. 106. ZwOlfer, H. 1965. . Preliminary l i s t of phytophagous insects attacking wild Cynareae (Compositae) i n Europe. Tech. Bull. Commonw. Inst. Biol. Control 6: 81-154. 107. Zwolfer, H. 1968. Some aspects of biological weed control i n Europe and North America. Proc. 9th Brit. Weed Control Conf. pp. 1147-1156. 108. Zwolfer, H. 1969. Experimental feeding ranges, of species of Chrysomelidae (Col.) associated with Cynareae (Compositae) in Europe. Tech. Bull. Commonw. Inst. Biol. Control 12: 115-130. •99 109. Zwolfer, H. 1970. Investigations of the natural enemies of some weeds. Rep. Commonw. Inst. Biol. Control. 1969. pp.. 34—37. 110. Zwolfer, H. 1970. Investigations on the host-specificity of Urophora aff i n i s Frfld. (Dipt.-, Trypetidae). Prog. Rep. Commonw.' Inst. Biol. Control No. 25: 1-28. 111. Zwolfer, H. 1971. Personal communication with P. Harris, May 24. 112. Zwolfer, H. 1971. . Weeds: Canadian Agricultural Projects. Quarterly Prog. Rep. Commonw. Inst. Biol. Control No. 2: 5-6. 113. Zwolfer, H., and P. Harris. 1971. Host specificity determination of insects for biological control of weeds. Ann.. Rev. Eht. 16: 159-178. VI. APPENDIX APPENDIX TABLE I. Effect of temperature on germination of C. diffusa and C. maculosat ,': Duration of Temperature (°C)tt r % test (days) 7 10 13 16 19 22 25 28 31 34 Linear Quadratic Residual One Oa • Oa Oa Oa 14a 41b 59b . 45b 12a 5a 21.04* 25.75* 53.19 Two (diffusa) Oa Oa 10a 57cd 86e 76.de lOOe 76de 48bc 24ab 18.42 17. 35 64.21 Two (maculosa) Oa Oa 43bc 86e 76de 95e 86e 90e 38bc 5a 14.80* 72.91* 12.28 Two (Av.)• Oa Oa 26bc 71d 81de 86de 93e 83de 43c 14ab 14.80* 72.91* 12.28 Three Oa 26b 74c 88c 88c 91c \"95c 86c 50 14ab 3.72* 92.50* 3.77 Four 10a 64bc 81cd 88d 88d 95d 98d 86cd 52b 19a 0.12 92.36* 6.51 Five 21a 76b 91b 88b 88b 95b 98b 86b 55 19a 0.97 88.48* 10.54 Six 24a 81b 91b • 88b 91b 95b 98b 86b 55 24a 1.25 87.20* 11.55 Seven 26a 81b 91b 88b 91b 95b . 98b 86b 55 24a 1.64 87.24* 11.10 Eight 33a 81b 91bc 91bc 91bc 95bc 100c 86bc 55 24a 3.07* 87.24* 9.68 Nine 33a 81b ' 91bc 91bc 91bc 95bc 100c 86bc 55 24a 3.07* 87.24* 9.68 Ten 33a 81b 91b 93b 91b 95b 100b 88b' 55 24a 2.88* 87.30* 9.81 t combined analysis for the two species except for day two when there was a significant (P = 0.05) treatment by species interaction. t t means i n the same row sharing the same letter were not significantly different at the 5% level according to Duncan's New Multiple Range Test. 2 *\" r which were significant at the 5% level according to Duncan's New Multiple Range Test. ,101 APPENDIX TABLE I I . Effect of Sowing Depth on Seedling Emergence of ' £ • diffusa, and'C. maculosa Species Sowing depths Duration of Test (Days) Eight Sixteen Twenty-Four Diffusa 0 cm 58 80 81 0.5 cm 58 63 62 1 cm 70 68 68 3 cm 3 16 16 5 cm 0 0 0 Maculosa 0 cm 58 90 91 0.5 cm 75 73 73 1 cm 70 78 78 3 cm 17 26 26 5 cm 0 7 8 "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0302441"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Plant Science"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Biology and control of centaurea diffusa lam. and centaurea maculosa lam"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/33706"@en .