HOST PLANT RESISTANCE TO WHITEFLY (TRIALEURODES VAPORARIORUM) IN THE GENUS LYCOPERSICON by Richard Ernest Vei l leux B.Sc. Tufts Univers i ty, 1970 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Dept. of Plant Science We accept th i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1976 (o) Richard Ernest Veilleux, 1976 In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th i s thes i s for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thes i s fo r f i nanc ia l gain sha l l not be allowed without my wr i t ten permission. Depa rtment The Univers i ty of B r i t i s h Columbia 2075 Wesbrook P l a c e V a n c o u v e r , C a n a d a V6T 1W5 i i ABSTRACT The greenhouse whitef ly (Trialeurodes vaporariorum) i s one of the most destructive pests of greenhouse tomato (Lycopersicon esculentum) crops. The success of b io log ica l or i n sec t i c ida l control of wh i te f l i e s has never been complete. Attention has therefore been directed recently to the p o s s i b i l i t y of developing tomato cu l t i va r s res i s tant to wh i t e f l i e s . Whitef l ies were caged on l ea f l e t s of ten tomato c u l t i v a r s , two of the woolly mutant l ines and four other species of Lycopersicon to observe the e f fec t of host plants within th i s genus on fecundity and longevity of the insects. The results showed wide var iat ion among hosts. S i gn i f i cant negative correlat ions were revealed between the mean density of glandular hairs on the upper f o l i a r surface of d i f fe rent cu l t i va r s and means for the fecundity of wh i te f l i e s caged on these cu l t i v a r s . A high level of r e s i s -tance to wh i te f l y , not related to density of glandular ha i r s , was observed in plants that were e i ther Van Wert's woolly mutant or L_. peruvianum var. humifusum. Resistance of the former seemed to be related to a high density of branched non-glandular trichornes whereas that of the l a t t e r was not morphologically apparent. There were indications of both ant ib ios i s and nonpreference operating in the humifusum.- Further experimentation revealed a high nymphal mortal i ty for wh i te f l i e s developing on plants of th i s l i n e , reduced fecundity of adults which had developed on the humifusum, and a preponderance i i i of male progeny from adult insects which had l i ved exc lus ive ly on these plants. I t was concluded that the resistance of L_, peruvianum var. humifusum to whitef ly i s s u f f i c i en t to j u s t i f y i t s use in a breeding program to develop greenhouse tomato cu l t i va r s res i s tant to th i s pest. iv TABLE OF CONTENTS page Introduction 1 L i terature Review 3 Whitefly biology 3 Damage to tomato crops 9 Pest resistance in Lycopersicon 12 Glandular hairs 18 Materials and Methods 21 General 21 Experiment I. Var ietal ef fects on whitef ly character i s t i c s in some cu l t i va r s of L_. esculentum 28 Experiment II. Duplication of Experiment I with methodological modif ication 30 Experiment I I I. Ef fect of pre-emergence fumigation on character i s t i c s of adult wh i te f l i e s 31 Experiment IV. Search for resistance to whitef ly in miscellaneous Lycopersicon 34 Experiment V. Comparative development of whitef ly on res i s tant and susceptible hosts 37 Experiment VI. Further studies of the ef fects of res i s tant host plants on whitef ly 38 Results 40 Experiment I 40 Experiment II 44 Experiment III 47 Experiment IV 50 Experiment V 56 Experiment VI 59 Discussion 63 Conclusion 80 L i terature Cited 81 Appendices 88 V LIST OF TABLES page Table I. Published data on fecundity of wh i te f l i e s . . . . 6 Table II. Source of plant material used in whitef ly resistance studies 21 Table I I I. Var ietal character i s t i c s of tomato cu l t i va r s studied in Experiment I . . 29 Table IV. The e f fect of tomato cu l t i v a r on means for whitef ly longevity, ov ipos i t ion rate and fecundity, and glandular hair density on both upper and lower leaf epidermis of the cu l t i va r s (Experiment I) 41 Table V. The e f fect of tomato cu l t i v a r on means for whitef ly longevity, ov ipos i t ion rate and fecundity, and glandular hair density on both upper and lower leaf epidermis of the cu l t i va r s (Experiment II) 45 Table VI. Means for longevity and ov ipos i t ion rate of wh i te f l i e s as affected by pre-emergence fumigation ( ' Su l fotep ' ) vs. control on two tomato cu l t i va r s (Experiment II I) 48 Table VII. Mean ov ipos i t ion rates obtained in Experiment III arranged according to the plant on which they were recorded . 48 Table VII I. The e f fect of L_. p impinel l i fo l ium and three tomato c u l t i v a r s , a l l of which appear in the pedigree of Swift , compared with the e f fect of Swift on means for longevity, ov ipos i t ion rate and fecundity of wh i t e f l i e s , and means for glandular hair density on e i ther f o l i a r epidermis of the plant material (Experiment IVa) 51 Table IX. Effect of some Lycopersicon accessions, tomato woolly mutants and greenhouse tomato cu l t i va r s on means for longevity, ov ipos i t ion rate and fecundity of wh i t e f l i e s . (Experiment IVb) 54 Table X. Mean density of glandular hairs on Vantage, F] (Vantage x L. hirsutum) and L_. hirsutum 55 Table XI. Comparative development of whitef ly on the susceptible tomato cv. Tiny Tim (TT) and the res i s tant accession of L_. peruvianum var. humifusum (Lph) (Experiment V) 57. vi LIST OF TABLES (Cont.) page Table XII. Comparative development of whtt.eflies whose parents spent various amounts of time on res i s tant (Lph) and susceptible (TT) host plants (Experiment Via) 60 Table XII I. Influence of susceptible (TT) and res i s tant (Lph) host plants on character i s t i c s of whitef ly adults whose pre-emergence development occurred on e i ther TT or Lph (Experiment VIb) 62 Table XIV. Comparison of whitef ly fecundity as measured in Experiments I and II and by Curry and Pimentel (1972) 72 v i i LIST OF FIGURES AND ILLUSTRATIONS page Plate 1. Leaf cage surrounding a secondary l e a f l e t of tomato cv. Campbell 1327 25 Figure 1. Design of Experiment III 33 Figure 2. Pedigree of Swift and var ie ta l descr ipt ion of plant material studied in Experiment IVa 35 Figure 3. Oviposit ion rate vs. age of female wh i te f l i e s . . 43 Figure 4. Development time (egg-adult) of wh i te f l i e s on two hosts 58 Plate 2. Portion of upper epidermal surface of a young l e a f l e t of Van Wert's woolly mutant 67 Plate 3. Portion of upper epidermal surface of a young l e a f l e t of tomato cv. Vendor . . . 68 Plate 4. Lycopersicon peruvianum var. humifusum 77 ACKNOWLEDGEMENTS I am grateful for the competent guidance of my advisor, Dr. C A . Hornby, the sound advice and constructive c r i t i c i s m of my committee members, Drs. H.R. McCarthy, J.H. Myers, C.W. Roberts and V.C. Runeckles and the r e l i ab l e support of my wi fe , Karen, throughout th i s work. I am also grateful for the s t a t i s t i c a l advice of Drs. G.W. Eaton and R.G. Petersoncand the photographic expertise of Dr. K. Bryant. 1 INTRODUCTION "Much evidence i s now accumulating of the existence of pools of genetic v a r i a b i l i t y in both plants and pests which would make possible the evolutionary modif ication of plant resistance and s u s c e p t i b i l i t y . " - - van: Emden/ 1966 Whitef l ies (Trialeurodes vaporariorum Westw.) have been a plague on greenhouse tomato (Lycopersicon esculentum M i l l . ) production for more than a century. These insects are s u f f i c i e n t l y tenacious and adaptable to have survived numerous attempts at i n sec t i c i da l and b io log ica l contro l . The search for resistance to wh i te f l i e s in the mu l t i -tude of tomato cu l t i v a r s , botanical va r ie t ie s of L,. esculentum, the genus Lycopersicon and c lose ly related Solanum species has only recently commenced. Although some potential sources of resistance have been found, th i s search is by no means complete. A breeding program has recently been launched (de Ponti et al_., 1975) with the primary objective of developing a whitef ly res i s tant tomato cu l t i v a r . The sever ity of the whitef ly control problem has been observed repeatedly in experimental work on Solanaceous crops in the UBC greenhouses. It was in response to interference by wh i te f l i e s with e a r l i e r , unrelated work in th i s greenhouse complex that current interest in the problem was generated. The objectives of the fol lowing study are: 1) to re-evaluate with what i s considered to be a more accurate methodology the levels of resistance to whi tef ly in tomato cu l t i va r s as 2 reported by Curry and Pimentel (1972);. 2) to determine i f the f o l i a r density of glandular ha i r s , a mechanism of insect resistance frequently c i ted in the l i t e r a t u r e , has an impact on whitef ly longevity or fecundity on the tomato cu l t i va r s studied. 3) to expand the search for resistance to whitef ly within the genus Lycopersicon to include plant material which has not been previously evaluated. 3 LITERATURE REVIEW WHITEFLIES Biology Greenhouse wh i te f l i e s (Trialeurodes vaporariorum Westwood) were f i r s t reported as pests on glasshouse plants in 1856 by Westwood who speculated that the species was indigenous to Mexico or B r a z i l . Russell (1948), however, claims that X* vaporariorum i s native to Southwestern North America because i t i s c lose ly re lated to other Trialeurodes species known to occur only in that area. In any case, J . vaporariorum i s now ubiquitous; Russell (1963) reports a worldwide d i s t r i bu t i on including every habitable continent and l i s t s hundreds of host plants. Whitefly anatomy has been described by Br i t ton (1902), Hargreaves (1915) and Russell (1948). Class-i f i c a t i o n of the Aleurodidae has been undertaken by Cockerel 1 (1902), Quaintance and Baker (1914 and 1917), Trehan (1940) and Russell (1948). Whitefly biology, l i f e history and habits have been studied by several authors: Westwood (1856), Br i t ton (1902), Mo r r i l l (1905), Hargreaves (1915), Williams (1917), S t o l l and Shull (1919), Schrader (1920 and 1926), Lloyd (1921 and 1922), Garman and Jewett (1922), Thomsen (1925), Speyer (1929), Weber (1931), M i l l i r o n (1940), Hussey and Gurney (1958 and 1960), and Scopes and Biggerstaff (1971). A br ie f account of the l i f e history of wh i te f l i e s and a composite of the i r habits fo l lows, with emphasis on fecundity and longevity. L i fecyc le Whitef l ies belong to the order Homoptera, family Aleurodidae. The females deposit eggs in c i r c l e s or arcs on the underside of young, succulent, 4 glabrous leaves, or scattered on pubescent leaves. Freshly emerged wh i te f l i e s mature within 48 h of emergence and females continue to ov ipos i t throughout the adult l i f e . The eggs are white at f i r s t but darken in about two days. There are three nymphal i n s ta r s , a pupal ins tar and f i n a l l y , the imago, or adult. The f i r s t nymphal i n s t a r , the only act ive one, moves a few mi l l imeters , inserts i t s s ty le t s into the phloem c e l l s of the vascular bundles and s e t t l e s , usually within 24 h of hatching. A l l subsequent development up to the adult stage occurs at the s i t e selected by the f i r s t nymphal i n s ta r , which i s incapable of moving o f f the leaf on which i t hatched. Development i s a function of temperature and photoperiod, so there i s considerable var ia t ion in the reports on the duration of the i n s ta r s , ranging from a 19.2-day l i f e cycle at 26.67°C (80°F) (Hussey and Gurney, 1958) to a 101-day l i f e cycle in a greenhouse between October and January, reported by Hargreaves (1915). Burnett ' s (1964) data are most probably typ ica l of operating greenhouses: egg 5-6 days (nymphal) instar I 5 days (nymphal) instar II 8 days (nymphal) ins tar III 2 days (pupal) instar IV 4 days to ta l 25 days Adults, which are less than 1.5 mm. long (Hargreaves, 1915), l i v e an average of 30 to 40 dctys (Garman and Jewett, 1922; Hargreaves, 1915; Gibson and Ross, 1940). Burnett (1949) measured longevity of females 5 as a function of temperature and found a peak of 50.5 days mean l i fespan at 15°C with longevity rapidly decl in ing as temperature diverged in e ither d i r ec t i on . Fecundity Reports on fecundity of wh i te f l i e s vary even more than those on dura-t ion of the l i f e cyc le. Table I presents published data on the da i l y ov ipos i t ion rate and the tota l eggs l a i d per female. Whitefly fecundity i s observed to be far from uniform, the var iat ion being influenced by a number of factor s . Of par t i cu la r interest i s the discrepancy between the reports of Burnett (1949) and Hussey and Gurney (1958). Burnett bases much of his experimentation on the "peak fecundity" observed at 18°C; ye t , i f ov ipos i t ion rate i s a f a i r estimator of tota l fecundity, then Hussey and Gurney's data indicate a greater fecundity at a much higher temperature (26.67°C, 80°F). Insect fecundity i s known to be influenced by a m u l t i p l i c i t y of factors . Engelmann (1970) discusses several of these including la rva l and adult n u t r i t i o n , age of female insects , number of ovarioles present, temperature, r e l a t i ve humidity and whether the insects are reared s ingly or in groups. Very l i t t l e work has been done to determine which of these factors influence whi tef ly fecundity most ser ious ly . Haseman (1946) reported that wh i te f l i e s were less attracted to plants grown on f u l l nutr ients than to those suffer ing from either phosphorus or magnesium def ic iency. Burnett (1949) studied whitef ly fecundity as a function of 6 Table I. Published data on fecundity of wh i t e f l i e s . Daily o v i -pos it ion rate eggs/V/day Total eggs/? Conditions Source 3-4 75 Garman & Jewett (1922) 3 130 Lloyd (1922) 5 130 Speyer (1929) 2 detached toibaceo Weber (1931) leaves 88 Gibson & Ross (1940) 2.18 ± 0.10 93.56 + 6.90 8.17 ± 0.23 319.58 + 17.73 8.39 ± 0.28 209.46 + 14.30 7.49 ± 0.28 123.90 + 0.09 8.10 ± 0.51 29.50 + 44440 4.04 ± 0.26 19.06 + 2.31 7.41 ± 0.21 79.70 + 22S60 2.52 ± 0.29 17.24 + ?2;23 2.1 to 10.3 0.8 to 5.2 4.2 ± 1.1 5.8 ± 3.4 10.8 ± 3.7 15°C 18°C 21°C 24°C 27°C 30°C 18°C (young leaf) 18°C (old leaf ) w , detached * ° H u n g tomato o l d leaf 15.6°C (60°F) • °FJ 23.9°C 26.7°C (75 c (80°F) 92.1 ± 11.8 249.7 ± 19.1 on tomato cv. Delicious on tomato cv. Tiny Tim Burnett (1949) detached tomato leaves in l ighted incubator Hussey & Gurney (1960) Hussey & Gurney (1958) Curry & Pimentel (1971) 7 temperature. Hussey and Gurney (1960) found that whitef ly fecundity decreased with increase in age of the tomato l e a f l e t on which the insects fed. These authors postulated that a wh i t e f l y ' s choice of feeding s i t e was determined by negative geotropism, color cues and the nut r i t i ona l status of a leaf as determined by feeding probes. When wh i te f l i e s were given a choice between feeding on a young or an old tomato l e a f l e t , a l l of them sett led on the young l e a f l e t within twenty minutes, even though many had i n i t i a l l y al ighted on the older l e a f l e t . In add i t ion, Hussey and Gurney reported that whitef ly fecundity decreased on phosphorus def i c ient p lants, regardless of the nut r i t i ona l status of the plants on which the nymphs developed. There was a s t a t i s t i c a l l y i n s i gn i f i c an t but evident decrease in fecundity i f the la rva l stages developed on phosphorus de f i c ient plants. Scopes and Biggerstaff (1971) reported that the fecundity of adult wh i te f l ie s was reduced by half i f they were kept without food in sealed glass tubes for more than two hours. Habits The adult insects seldom f l y unless disturbed (Speyer, 1929). Hargreaves (1915) considered that whitef ly adults were neither pos i t i ve l y nor negatively phototropic, but Lloyd (1921) observed that wh i te f l i e s are attracted to yel low, and w i l l choose a yellow card rather than a tomato plant i f both are equidistant from an i n fe s ta t i on . M i l l i r o n (1940) also found evidence suggesting a phototropic response and Weber (1931) reported that whitef ly females are guided by intens i ty rather than wave-length of l i g h t for se lect ion of ov ipos i t ion s i t e s . Garman and Jewett (1922) noted that i so lated pairs of wh i te f l i e s did not mate read i l y ; Lloyd 8 (1922) observed the gregariousness of adult wh i t e f l i e s , because massive infestat ions were often loca l i zed on a few plants whereas nearby plants were clean or nearly so. Garman and Jewett (1922) did not observe female wh i te f l i e s to have copulated more than once but Lloyd (1922) observed copu-la t ion between the same pair several times over a period of several weeks. The pubescence of the leaves on which wh i te f l i e s had developed was found to influence the morphology of pupal and adult stages of the insect (Russel l , 1948). Adu^ts-whdch^hadadevel 'opedednoglabrbusL-.:!eaves were ' larger thanzadulitsrwhcich hadhdeye.ilibpedironjipuBescent^leav.esv" \-Lloyd': (1922) observed a high nymphal mortal i ty on plants with a th ick c u t i c l e . Parthenogenesis Hargreaves (1915), working in Great B r i t a i n , reported that un f e r t i l i z ed females produced a l l female progeny parthenogenetically whereas. .Garman a'nd Jewett (1922) and Lloyd (1922), working independently in Kentucky and Great B r i t a i n respect ive ly , reported that un fe r t i l i z ed females produced a l l males parthenogenetically. This led to the be l ie f that two races of whitef ly ex i s ted; Schrader (1920), however, could not f ind representatives of the B r i t i s h race (producing females only) while conducting an invest iga-t ion in the same area where Hargreaves had worked. He concluded that the American race was possibly spreading and supplanting the B r i t i s h race. Williams (1917) suggested that f e r t i l i z e d eggs of the American race of wh i te f l i e s develop into either males or females, but S t o l l and Shull (1919) produced evidence from breeding experiments that f e r t i l i z e d eggs of the American race develop into females and un fe r t i l i z ed eggs develop 9 into males (arrhenotokous parthenogenesis). Schrader (1920) confirmed th i s work with cyto log ica l evidence and concluded that f e r t i l i z a t i o n of eggs in a mated female i s under the control of the female and depends upon an unident i f ied stimulus. Thomsen (1925), who worked with both the English and American races,suggested that rather than two races, there might ex i s t two types of female: 1) ob l i g a t o r i l y parthenogenetic thelytokous, corresponding to the so-cal led English race, and 2) f a cu l t a t i ve l y parthenogenetic arrhenotokous, corresponding to the so-cal led American race. Haploid males (n=l l , Schrader, 1920) occur in both " races" . They are extremely rare in populations with thelytokous parthenogenetic females, but occur in a ra t io of about 50:50 in populations with arrhenotokous parthenogenetic females (Garman and Jewett, 1922; Speyer, 1929). Damage to tomatoes Three categories of damage by wh i te f l i e s to glasshouse crops are dist inguished: 1) d i rec t feeding i n ju ry , 2) honeydew secretions and 3) disease transmission. Foliage feeding by adults and so-cal led scales ( i . e . nymphal and pupal instars) "weaken growth and induce w i l t i n g " of tomato plants (Hussey et al_., 1959). Br i t ton (1902) states that "the t issues collapse from the effects of th i s continuous pumping out of the l i f e ju ices of the p lant, and the leaf shr ive l s and f a l l s . " S imi lar descr ipt ions of feeding damage can be found.in many of the references c i t e d . Lindquist 10 et_ al_. (1972) document the ef fect of d i f fe rent population dens i t ies of wh i te f l i e s on the y i e l d of greenhouse tomato crops. They found that high density populations reduced y i e l d , espec ia l ly late in the season. Hussey et al_. (1959) state that the pr inc ipa l injury that wh i te f l i e s cause to tomato crops results from the excretion of honeydew, which interferes with resp i rat ion by clogging stomata, cand reduces photosynthesis by obscuring the leaf surface. Honeydew i s the excrement of wh i t e f l i e s , produced most copiously during the pupal in s ta r . This sugary excrement promotes the growth of a sooty mould, commonly Cladosporium sphaerospermum Penzig,J;t"he spores of which germinate at RH 90%, a-eondiition. often met in glasshouses. Growth of th i s mould on leaves and f r u i t , which occurs even in mild whitef ly in fes ta t ions , further i nh ib i t s photosynthesis and often makes the f r u i t unmarketable without an expensive cleaning operation. Whitef l ies have also been found to be vectors of several pers istent or semi persistent viruses (Varma, 1963). Duffus (1965) de ta i l s the mode of transmission of beet pseudo-yellows virus by wh i t e f l i e s . In addit ion, Bugbee (1962) has found that wh i te f l i e s transmit a bacter ia l disease, Xanthomonas pelargonii (Brown) on geraniums. Methods of control The l i t e r a t u r e on i n sec t i c i da l control of wh i te f l i e s i s extensive; of the authors previously mentioned, Westwood (1856), Br i t ton (1902), Mo r r i l l (1905), Lloyd (1922), Garman and Jewett (1922) and Anon. (1973) a l l give recommendations for chemical contro ls . Current pest control handbooks 11 provide recent recommendations. The problem for more than a century has been that no in sect i c ide i s tox ic to a l l stages in the l i f e cyc le , which thereby prevents eradicat ion. The egg stage i s res i s tant to hialathion, Perthane, maneb and zineb (McMullen, 1964). In order for i n sec t i c i da l control to be e f f e c t i v e , fumigation at short regular interva l s i s necessary, a method both eco log ica l l y and economically expensive. Adult resistance to insect ic ides in stra ins of whitef ly has been reported by Wardlow et^ al_. (1972) (DDT, malathion, and dichlorvos) and Parr et a l . (197-5) (resmethrin). Paras i t i za t ion of whitef ly nymphal instars by the chalc id wasp, Encarsia formosa Gahan, has been known since 1924 (McLeod, 1938). Speyer (1927) f u l l y described the mode of pa ra s i t i za t i on . Several attempts have been made to establ i sh e f fec t i ve b io log ica l control programs employing this t iny parthenogenetic wasp. Several factor s , pa r t i cu l a r l y temperature (McLeod, 1938; Mi 11 i ron , 1940; Burnett, 1949) must be ca re fu l l y contro l led in order to establ i sh a successful b io log ica l control program. The system is also sens it ive to photoperiod, RH, leaf pubescence ( M i l l i r o n , 1940) and i n i t i a l d i s t r i bu t i on of the host population (Parr, 1968). The synchronization of host and parasite release i s also very important (Parr, 1970; Scopes and Biggerstaff, 1971). McClanahan (1970) outl ines an integrated control program with Encarsia formosa and the multi-purpose pest ic ide, oxythioquinox (Morestan), which i s tox ic to Trialeurodes but 12 not to Encarsia. However, th i s chemical has not been registered for use on tomatoes. Although the use of Encarsia has been demonstrated to be e f fect i ve under carefu l l y control led condit ions, i t s widespread acceptance seems un l ike ly because of i t s s en s i t i v i t y to environmental condit ions. In another form of b io log ica l con t ro l , Hussey (1958) attempted to use a paras i t i c fungus, Cephalosporium aphidicola Petch, which attacks the nymphal and pupal instars and adult wh i t e f l i e s . Its spores, however, require saturation in water for 24 hours for sat i s factory germination, thus making i t s use impractical fo r greenhouse management. MacDowal (1972) suggested the p o s s i b i l i t y of whitef ly control by a t t rac t ing them to l i g h t traps. This idea i s s t i l l speculative and would involve much more research before i t s r e a l i z a t i on . Pest resistance in Lycopersicon and re lated species. The search for insect resistance in tomatoes and related species i s recent; nearly a l l such research has been done within the l a s t ten years. Maxwell et al_. (1972) published a par t i a l review of the subject. No tomato cu l t i v a r has yet been released which i s the product of a breeding program in which insect resistance was the primary object ive, although several plant breeders are current ly involved in such programs (de Ponti ejt al_., 1975; Clayberg and Kring, 1974). A b r i e f review follows of plant resistance to insects which has been found within the genus Lycopersicon and c lose ly related Solanum species. 13 Leaf-feeding pests Whitef ly, Trialeurodes vaporariorum Westwood Hussey and Gurney (1960) measured the fecundity of whitef ly on three greenhouse tomato c u l t i v a r s , but found no s i gn i f i can t d i f ferences. Genti le et al_. (1968) v i s ua l l y evaluated the resistance to whitef ly of 16 Lycopersicon and Solanum accessions and three tomato c u l t i v a r s . A l l were susceptible except two pa r t i cu l a r l y hirsute plants of L_. hirsutum Humb. and Bonpl. and a l l of the penne l l i i Corre l l plants tested. If the glandular hairs were brushed from the fo l iage of e ither accession and wh i te f l i e s were allowed to ov ipos i t , development proceeded normally from egg to adult. Curry and Pimentel (1971) measured longevity and fecundity of adult female wh i te f l i e s caged on l ea f l e t s of both tomato cvs. Tiny Tim and Del ic ious. They found both mean l i fespan and mean fecundity to be s i g n i f i c an t l y greater on Tiny Tim.but there were no s i gn i f i can t differences in rate of development or nymphal mortal i ty on the two cu l t i v a r s . Curry and Pimentel (1972) screened 90 tomato cu l t i v a r s for resistance to whitef ly and further evaluated 12 which had been designated as comparatively res i s tant or susceptible. F i f t y adult wh i te f l i e s were caged on each of eight indiv idual plants of the 12 cu l t i va r s and the number of adults/cage was counted 42 days l a t e r . A spectrum of resistance was indicated; however, the highly s i gn i f i can t differences between whitef ly performance on Tiny Tim and Delicious reported in t he i r 1971 paper were not confirmed. Hogenboom et al_. (1974) and de Ponti • et al_. (1975) have reported resistance to whitef ly in accessions of L_. hirsutum, I. hirsutum f. 14 glabratum Mul 1., L_. p impine l l i fo l ium Mi 11., L_. chilense Dun., L_. esculentum cv. Heinz 1370 and Solanum penne l I i i . The mechanism of glandular hairs has been evoked only with respect to S. penne l ! i i . Clayberg and Kring (1974) demonstrated that the resistance to whi tef ly of S_. pennel! i i i s eas i l y transferred to tomatoes. Clayberg (1975) developed a p e r i c l i n a l plant chimera composed of core t issue of J_. esculentum with the epidermis of S_. pennel I i i . This plant was as res i s tant to whi tef ly as S_. pennel! i i i t s e l f . Clayberg then concluded that st ick iness caused by the exudate of glandular hairs i s the major, i f not the only mechanism of resistance in S_. pennel I i i . In a study of the resistance of S. penne l ! i i , Plage (1975) drew the fol lowing conclusions: There i s high whitef ly nymphal morta l i ty on S_. penne l ! i i . There i s a corre lat ion coe f f i c i en t of r = -0.813 between levels of whitef ly in festat ion and st ick iness of fo l i age . H e r i t a b i l i t y of resistance = 0.75. The F, of a susceptible tomato cu l t i v a r x S_. pennel1ii i s midparental in res istance. There are fewer res i s tant plants than expected in l a te r generations, possibly due to linkage with l e tha l s . A level of s t ick iness below that of S_. pennel! i i might s t i l l be valuable in cont ro l l i ng wh i t e f l i e s . Aphids, Myzus persicae (Sulz.) Macrosiphum euphorbiae (Thomas) Aphis craccivora (Koch) The l i nk between the exudate of glandular hairs on tomato fo l iage and insect ( s p e c i f i c a l l y , Myzus persicae) resistance was f i r s t suggested by McKinney (1938). Johnson (1956) confirmed McKinney's report with 15 respect to M. persicae and reported a s im i l a r resistance to Aphis cracc ivora. He suggested that the glandular hair exudate contained a tox ic p r inc ip le in addit ion to i t s obvious mechanical e f fec t . Stoner et a l . (1968|))and Genti le and Stoner (1968a) screened several hundred tomato cu l t i v a r s , breeding l i ne s , and accessions of Lycopersicon and Solanum for resistance to Macrosiphum euphorbiae. They found high levels of resistance in f i v e of 21 accessions of J_. peruvianum M i l l , and immunity in Solanum p e n n e l l i i . The mechanisms for resistance were attr ibuted to a "physiological incompat ib i l i t y " in L.. peruvianum and to exudate from glandular hairs in S.« p e n n e l l i i . The resistance of S_. penne l l i i to M. euphorbiae was further explored by Clayberg and Kring (1974), who found that i t was contro l led by r e l a t i v e l y few genes and could be eas i l y transferred to tomato c u l t i v a r s . Clayberg (1975), employing the same p e r i c l i n a l plant chimera used in studying whitef ly res istance, concluded that glandular hair density was not the only component of aphid resistance in S_. penne l l i i but that a, component of the c e l l sap was involved. Other leaf-feeding insects, Liriomyza munda (Fr ick) Ep i t r i x h i r t ipenn i s (Melsheimer) Wolfenbarger (1966) found varying levels of resistance to leaf miners, Liriomyza munda and the tobacco f l e a beet le, Ep i t r i x h i r t ipennis in tomato cu l t i va r s and tomato mutant stocks. The woolly tomato mutant was the least injured by leaf miners. Genti le and Stoner (1968b) continued evaluating tomato cu l t i va r s and accessions of Lycopersicon for resistance to tobacco f l ea beetle. Although a l l were susceptible to la rva l feeding, young leaves of accessions of L_. hirsutum and'L. hirsutum f. glabratum 16 were undamaged by adults. This was again ascribed to a high density of glandular ha i rs . Webb et al_. (1971) confirmed that these same accessions were v i r t u a l l y immune to the tobacco f l ea beetle. Although L_. pimpinel 1 i -folium appeared to have a promising a n t i b i o t i c mechanism to tobacco f l e a beetles in greenhouse te s t s , f i e l d tests were disappointing. Mi'tgs» Tetranychus urt icae (Koch) Tetranychus te la r iu s (Linn.) Tetranychus marianae (McGregor) Tetranychus cinnabarinus (Boisduval) Wolfenbarger (1965) f i r s t observed incomplete mite (T. marianae) resistance in tomato plants carrying the woolly gene and also in three Lycopersicon accessions. The tomato cvs., Anahu and Kalohi, which were selected for disease resistance a f te r incorporation of genes from J_. p impine l l i fo l ium, L_. peruvianum and L_. hirsutum, have received much attention with respect to mite res istance. G i lber t et al_. (1966) and Stoner and Str ingfel low (1967) have found that the tolerance of these cu l t i va r s to feeding by X* te la r iu s seems to have a complex pattern of inheritance, evidenced by the intermediate tolerance of hybrids between e i ther Anahu or Kalohi and susceptible c u l t i v a r s . Stoner et a l . (1968a and Stoner (1970) l inked the resistance to X- cinnabarinus of these two cu l t i va r s to a high density of glandular hairs and successful ly selected for spider mite resistance on the basis of th i s high density. They suggested that th;is was not the sole resistance mechanism, however. Stoner and Gentile (1968), Genti le et aX. (1969) and Rodriguez et a l . (1972), while screening hundreds of Lycopersicon accessions for mite res istance, described a type of resistance duernto"glandular hairs in 17 accessions of J_. hirsutum, L_. hirsutum f . glabratum and Solanum penne l l i i • A s i g n i f i c an t negative cor re lat ion was found between glandular ha i r density and mite fecundity. Aina et al_. (1972) demonstrated the presence of a toxin letha l to mites in the glandular hair exudate of Anahu and Kalohi and that of a L.hhjrsutum accession. Cantelo et al_. (1974) found that ethanol extracts of glandular ha i r s , f o l i age , or f r u i t of the tomato plant contained materials tox ic to adult females and immature forms, but not to adult males, of T. urt icae and T. cinnabarinus. Fruit-feeding pests F r u i t f l y , Drosophila melanogaster Merg. Mason et al_. (1960) found evidence of va r ie ta l resistance to Drosophila. Stoner and Mason (1966 and 1969) and Stoner et al_. (1969 and 1972) ref ined the techniques for determining resistance in several tomato c u l t i v a r s , but s t i l l reported much var iat ion in resistance within a c u l t i v a r and even among d i f fe rent f r u i t s from a s ingle plant. No mechanism was proposed to explain the preference of Drosophila for ov ipos i t ion on f r u i t of some cu l t i v a r s , but the p o s s i b i l i t y of se lect ion for resistance to Drosophila was suggested. Tomato fruitworm, Hel iothis zea (Boddie) Canerday et/al_. (1969) reported d i f f e r i n g degrees of fruitworm injury to tomato cu l t i v a r s . A ser ies of papers by Fery and Cuthbert (1973, 1974a, 1974b and 1975) demonstrated a pos i t ive corre lat ion between fruitworm 18 damage, earl iness of the tomato cu l t i v a r and increasing plant density, but showed a negative corre lat ion between damage and vine s i ze . The authors confirmed the existence of an an t i b i o t i c factor to tomato fruitworm in accessions of L_. hirsutum f. glabra turn. This factor d id not appear to be unique but i t was at a higher level than i t was in _L. esculentum. An F-j hybrid between the two species was completely susceptible to fruitworm in ju ry , ind icat ing that the resistance was probably recessive. GLANDULAR HAIRS 1. Morphology Luckwil l (1943) described seven types of trichomes which occur in the genus Lycopersicon, three of which are glandular. His descriptions of the three glandular types fo l low: P.- Slender hairs;]0.2 to 0.4 mm. long; [4- to 8 -ce l led ; standing on a large simple basal c e l l , ] and with a small glandular ves ic le at the t i p . Occur only in L_. hirsutum. l2.^]Glandular ha i r s ; 0.1 to 0.5 rnrn. long; consist ing of a large un i ce l l u l a r base, a sta lk of 1 to 2 c e l l s , and a glandular head of 2 to 4 c e l l s . These glandular c e l l s secrete an o i l which accumulates beneath the c u t i c l e , distending i t un t i l the head i s 0.05 to 0.08 mm. in diameter. When the plant i s handled, the cu t i c l e breaks and the o i l , responsible for the characte r i s t i c scent of the p lant, i s l i be ra ted. This type of glandular hair occurs in a l l speciesoof Lycopersicon. [i3;]Sman glandular ha i r s ; 0.05 to 0.10 mm. long; consist ing of a s ingle basal c e l l , a un i ce l l u l a r s t a l k , and a head, 0.02 to 0.04 mm. in diameter, consist ing of 4 to 8 c e l l s i r r egu l a r l y arranged. Occur in a l l species except I. p impine l l i fo l ium. Uphof (1962) presents in elaborate deta i l the morphology and ontogeny of trichomes occurring throughout the plant kingdom. 19 Function A case for plant defense as the major function of glandular trichomes has been defended s k i l l f u l l y by Levin (1973). Gibson (1971) describes the process by which aphids (Myzus persicae and Macrosiphum euphorbiae) gradually become immobilized by accumulation on the insects ' t a r s i of exudate from the glandular hairs of three Solanum species. These hairs as well as nearly a l l those mentioned in insect resistance studies c i ted e a r l i e r , are comparable to type 2 in the c i t a t i o n from Luckwill (1943). Chemical constituents^ No work appears to have been done to ident i f y the chemical structures of substances present in the glandular hairs of any Lycopersicon species. Gibson (1971) presented evidence for the presence of phenolic compounds in the hairs of three Solanum species. Thurston et al_. (1966) reported the presence of two a l ka lo id s , n icot ine and anabasine, in trichomes of seven Nicotiana- species. Beckman ejt al_. (1972) presented the only paper dealing s p e c i f i c a l l y with the chemical composition of 4-lobed glandular hairs of tomato. They reported the occurrence of phenolic substances, and suggested that these hairs may serve the plant as sensory organs. Tobacco mosaic virus has been extracted from the glandular exudate of sys-te'mgca-l;;l>y;jnfected^toiuaio-:p.l,a.o$spCfelaxiJissand-*EJradJeys 1973).. The author claimed that th i s i s the prime, i f not the only in fect ion s i t e in aphid transmission of the virus and warned plant breeders of the p o s s i b i l i t y of increased transmission of virus diseases on insect res i s tant selections which have a 20 high density of glandular hairs . The major points in the l i t e r a tu re review are summarized as fo l lows. Whitef l ies are a tenacious pest of greenhouse tomato crops. No attempt at i n sec t i c i da l or b io log ica l control of wh i te f l i e s has been completely e f f e c t i v e . Insect resistance in the genus Lycopersicon has been only cur sor i l y explored and l i t t l e employed. That Which has been found has frequently been associated with a high density! of glandular hairs on the fo l i age . L i t t l e i s know of the chemistry of these glandular hairs or that of the i r s t i cky exudate. Whitefly fecundity and longevity are sens i t ive to a var iety of influences. If measured under ca re fu l l y control led condit ions, estimates of these parameters should r e f l e c t differences inmost plant acceptab i l i t y to the insect. 21 MATERIALS AND METHODS Source and handling of plants Seed for the d i ve r s i t y of plant material employed in the fol lowing study was obtained from the various sources l i s t e d in Table II. Table II. Source of plant material used in whitef ly resistance studies. Cu l t i va r , accession or mutant stock Source Campbell 1327 : Swift :-'Tiny Tirn Stokes Seeds L t d . , St. Catherines, Ont. :Roma VF v Ferry-Morse Seed Co. ( Inc.) , P?0. Box 100, Mountain View, C a l i f . 'Sunray ;- Dominion Seed House, Georgetown, Ont. 'Redskin 1 ' Farthest North Bounty . Research Stat ion, Canada Agr i cu l tu re , Morden, Man. L. p impine l l i fo l ium P.I.270444 L. peruvianum var. humifusum P.I. 127828 L. hirsutum P.I. 127826 Plant Genetics & Germplasm Inst., Agr icu l tura l Research C t r . , B e l t s v i l l e , Md. L. cheesemanii P.I. 231257 James L. J a r v i s , Regional Plant Introduction S ta . , Ames, l a . Vantage "Vendor . F-| ['.Vantage: x L. hirsutum] U.B.C. stocks. Morgan's Woolly Van Wert's Woolly CM. Rick, Dept. of Vegetable Crops, U. of Ca l i f o r n i a , Davis, C a l i f . 22 The seed was planted d i r e c t l y into 7.5- in. p l a s t i c pots containing a 3:1 mixture of s team-ster i l i zed soil:sphagnum moss, supplemented with 'Osmocote' (14-14-14). The plants were approximately four weeks old at the beginning of each experiment and were replaced a f te r an addit ional four weeks with new four-week old plants. For the duration of an exper i -ment, the plants were grown in a ' P e r c i v a l ' growth chamber adjusted to a 12 h photoperiod, a daytime temperature of 23.9°C (75°F) and a nighttime temperature of 20.0°C (68°F). After placement in the growth chamber, the plants were given approximately 50 ml. of a complete nutr ient solut ion (solution no. 2, Hoagland and Arnon, 1938) three times weekly. Source of wh i te f l i e s Whitef l ies were taken from a reservoir kept on tomato plants growing in an unfumigated greenhouse at the Univers ity of B r i t i s h Columbia. This house was never the d i rec t target of insect ic ides but was connected to a greenhouse complex that was fumigated about once monthly with one of Parathion, Pyrethrum, DDVP, or Plantfume 103; low levels of fumigants most cer ta in ly d r i f t ed into the unfumigated greenhouse on these occasions. This population of wh i te f l i e s had been present inland around the UBC greenhouses for a number of years. They were supported on a var iety of compatible, pa r t i cu l a r l y Solanaceous hosts throughout the year and defied erad icat ion. The genetic v a r i a b i l i t y in the whitef ly population was not studied in th i s research. Specimens from this population were pos i t i ve l y i den t i f i ed as Trialeurodes vaporariorum (Westwood) by W.R. Richards of the Biosystematics Research I n s t i tu te , Ottawa. 23 Cage Construction Several cage models were tested over a period of eight weeks before the fol lowing model was accepted. Reasons for f a i l u r e of e a r l i e r models were: t o x i c i t y of mater ia l s , e.g. residual vapors from the epoxy; imperfect seal ing methods which allowed wh i te f l i e s to escape; and too l i t t l e a l low-ance for a i r c i r cu l a t i on which caused a physiological disorder, oedema \ on the tomato l ea f l e t s within the cages. Cages of the most successful model were constructed of 8 x 11 cm. sheets of Kodacel into which four rectangular windows had been cut for ven t i l a t i on . Fine nylon mesh was then machine-sewn over the windows, the sheet of Kodacel was r o l l ed into a cyl inder of 3 cm. diam. and fused, em-ploying a narrow f i l m of Kodacel softened in acetone. A disc of poly-urethane foam 0.5 cm. thick and 3.5 cm. diam. into which a 1 cm. long s l i t had been cut, was then glued onto one end of the cyl inder using LePages 5-minute-epoxy. A s im i la r polyurethane d i s c , 1 cm. th i ck , into which a radia l s l i t had been cut, capped the opposite end of the cy l inder . The cages and caps were washed in 'Alconox' and dried for 12 hours in a 51.7°C (125°F) vent i lated oven to evaporate any residual toxins from the epoxy. Complete cages weighed 2.5-3.0 gm. Small wart l ike excrescences . . . on [the] underside of leaves . . . When roots take up more water than i s given o f f by leaves, the pressure . . . may cause enlarged mesophyll c e l l s to push outward through the epidermis." Wescott (I960) p. 283. A ce l lu lose t r iacetate f i l m developed by Eastman Chemicals, ava i lab le in Canada through P l a s t i c and Paper Sales, L td . , 140 Sunrise Ave., Toronto, Ont. M4A 1B3. 24 Management of wh i te f l i e s on the plants Leaf lets which were not quite f u l l y mature were chosen as s i tes for the cages. In natural ly infested plants, these leaves support most of the adult whitef ly populations and are therefore the s i t e of greatest egg deposit ion. The cages were placed on the plants.by f i t t i n g the pet io le of a secondary l e a f l e t into the s l i t of the cap and, i f necessary, removing any minute t e r t i a r y l e a f l e t s . The cage then surrounded the secondary l e a f l e t and was supported by the tension of the th ick polyurethane cap through which the pet io le extended, against the Kodaceil; cy l inder . (Plate 1). The day before freshly emerged wh i te f l i e s were needed, a l l adult wh i te f l i e s were removed from undetached tomato leaves bearing a dense population of pupal in s ta r s . Such leaves were wrapped in cheesecloth, and on the fol lowing day, a l l wh i te f l i e s present on the leaf were assumed to be less than 24 h o ld . Unless otherwise spec i f i ed , ten f resh ly emerged wh i te f l ie s ( f ive male and f i ve female) were placed in each Kodacel- cage surrounding a newly maturing l e a f l e t . Whitef l ies were handled with a mouth-suction asp i rator , made from a length of 0.64 cm. (0.25 in . ) rubber tubing, the end of which was covered with a small piece of s i l k mesh and f i t t e d into a length of 0.64 cm. (0.25 in. ) i d glass tubing which had been drawn to a point approximately 2 mm. diam. at the open end. The wh i te f l i e s could be sucked gently into the glass tube, sexed under a Cenco binocular d issect ing microscope (lOx), and blown gently into the cages by s l ipp ing the t i p of the aspirator through the s l i t in the polyurethane disc at the apical end of the l e a f l e t . 25 P l a t e 1. L e a f cage s u r r o u n d i n g a s e c o n d a r y l e a f l e t o f tomato cv. Campbell 1327. W h i t e f l i e s can be seen on t h e l e a f l e t w i t h i n the cage. 26 Co l lect ion of Data a. Longevity of wh i t e f l i e s . The wh i te f l i e s were changed to fresh l e a f l e t s at 5-to 7-day i n te rva l s . At each change, the encaged l e a f l e t was removed from the plant. Surviving wh i te f l i e s were co l lected in an asp i rator , counted and sexed. Whitefly eggs on the l e a f l e t were counted under a Cenco binocular d issect ing microscope (lOx). The empty cage was rinsed with alcohol and placed on a fresh l e a f l e t . The wh i te f l i e s were returned to the cage. Longevity values were recorded each time the cage was moved only for insects for which corpses were found. These insects were assumed to have died midway between the previous and the current count. I f wh i te f l i e s were missing, they were assumed to have escaped midway between the two counts. Occasionally, wh i te f l i e s were found which were f i rmly stuck to and struggling to free themselves from the sta lk of a glandular hair on a tomato plant; these insects could not be removed with an aspirator without damaging them. Such insects were considered dead on the day that the cage was moved. J u s t i f i c a t i o n for th i s assumption came from having found several corpses ( i f female, often surrounded by a large number of eggs) stuck in a s im i l a r manner and i t appeared that the wh i te f l i e s could not survive in th i s condition for more than a couple of days. There was a maximum of ten longevity values per cage. b. Oviposit ion rate for wh i t e f l i e s . The ov ipos i t ion rate for the f i v e female wh i te f l i e s in each cage was 27 computed by taking the to ta l number of eggs deposited within a cage for a designated period and d iv id ing that value by the tota l female days spent in that cage for the same period. In the f i r s t experiments, the designated period was the l i fespan of the longest- l ived female; l a t e r , th i s period was the f i r s t 21 days of each experiment. c. Fecundity of wh i t e f l i e s . The fecundity of the f i v e female wh i te f l i e s in each cage was computed by mult ip ly ing the da i l y ov ipos i t ion rate of the same f i v e females by the i r mean longevity. The mean da i l y ov ipos i t ion rate and fecundity for wh i te f l i e s on each tomato cu l t i v a r in an experiment was the mean of a l l the cage means for each t r a i t respect ively on that c u l t i v a r . Thus, da i l y ov ipos i t ion rate estimated the approximate number of eggs deposited by a s ingle female whitef ly each day on plants of a par t i cu la r tomato c u l t i v a r . Likewise, to ta l fecundity estimated the to ta l number of eggs deposited by a s ingle female during her l i fespan on plants of a par t i cu la r tomato c u l t i v a r . d. Glandular hair density of host plants. The density of glandular hairs was determined on l e a f l e t s opposite newly caged ones. Such l e a f l e t s were bisected and each half was used to estimate the glandular hair density on e i ther the upper or the lower epidermis. The bisected l e a f l e t was placed on a glass s l i de to which a f i lm of glycerol had been appl ied. The leaf surface was scanned under a Nikon monocular microscope (5x) fo r an appropriate s i t e , i . e . one which 28 d i d not i n c l ude any major ve ins and was f a i r l y c l o s e to the m id r i b a t the basal end of the l e a f . A l l g l andu l a r h a i r s appear ing i n the m i c r o -scop ic f i e l d (50x) were counted. The area o f the c i r c u l a r f i e l d was s l i g h t l y under f ou r square m i l l i m e t e r s ; h a i r counts are thus expressed 2 i n terms of the number of g l andu la r hairs/4mm . Th i s method was mod i f i ed from tha t desc r ibed by Stoner et al_. (1968a). Mean den s i t y of g l andu la r ha i r s was based on 15 or 17 counts on any of s i x p l an t s of each c u l t i v a r i n an exper iment. For each of these counts , made a t va r i ou s t imes over the course of an exper iment, a g l andu l a r h a i r den s i t y was determined f o r both upper and lower l e a f epidermis of every c u l t i v a r f o r which g l andu l a r h a i r den s i t y was de s i r ed i n t h a t experiment. For convenience, g l andu l a r h a i r d e n s i t i e s on the upper and lower l e a f ep idermis are abbrev i a ted i n the t a b l e s to GHD-UE and GHD-LE r e s p e c t i v e l y . Data were analyzed accord ing to the s t a t i s t i c a l procedures des6ribed9tDy)Zar (1974). ! Experiment I The f i r s t experiment was designed to r e - e v a l u a t e the r e s u l t s repor ted by Curry and Pimentel (1972). They found a spectrum of s u s c e p t i b i l i t y to w h i t e f l i e s i n twelve tomato c u l t i v a r s by caging 50 a d u l t w h i t e f l i e s on i n d i v i d u a l p l an t s and count ing the adu l t s per cage 42 days l a t e r . F i ve of these twelve tomato c u l t i v a r s , rep re sent ing the f u l l spectrum of s u s c e p t i b i l i t y to w h i t e f l y , were s e l e c ted to study v a r i e t a l e f f e c t on l ongev i t y and f e c u n d i t y of w h i t e f l i e s i n the present 29 study. The procedure for estimating these parameters, as previously out l ined, was s imi la r to that used by Curry and Pimentel (1971), and appeared to be more sens i t ive to var ie ta l e f fect on wh i te f l i e s than the mass in fes tat ion techniques employed by Curry and Pimentel (1972) and Genti le et a]_. (1968). A br ie f descr ipt ion of the characte r i s t i c s of the f i ve cu l t i va r s i s given in Table I I I. The density of glandular hairs was also studied in the f i v e cu l t i va r s to determine the degree of cor re la t ion between density (on e i ther upper or Tower lea f epidermis) and longevity, ov ipos i t ion rate or overal l fecundity in the wh i t e f l i e s . Table I I I. Var ieta l character i s t i c s of the tomato cu l t i va r s studied in Experiment I. Cu l t ivar Character i s t ics Roma VF semi-determinate p lant, l a te y i e l d , uniform r ipening, large plum-shaped red f r u i t , res i s tant to V e r t i c i l l i u m and Fusarium (Minges, 1972). Sunray indeterminate plant, la te y i e l d , large round orange f r u i t , re s i s tant to Fusarium (Minges, 1972). Tiny Tim determinate dwarf plant, early y i e l d , small red f r u i t , often grown as a houseplant (Minges, 1972). Swift determinate p lant, ear ly y i e l d , uniform r ipening, medium-sized dark red f r u i t (Blakely and Carl berg 1964) Campbell 1327 semi-determinate plant, la te y i e l d , large' s i i g h t l y f lattened red f r u i t , bred for processing, to lerant to V e r t i c i l l i u m and Fusarium (Stokes Seed Catalogue, 1975}: 30 Plants were seeded June 1, 1975 and grown in an insect - f ree green-house un t i l July 1, 1975 when they were transferred to the growth chamber. The f resh ly emerged wh i te f l i e s were taken from cv. Vendor or (Vendor x Tropic) tomato plants. Three rep l i cat ions were set up on ten plants (two of each c u l t i v a r ) . Each rep l i ca te consisted of f i v e leaf cages (one per c u l t i v a r ) containing ten whitef l ies/cage. Experiment II Experiment II was intended to ve r i f y the resu l t s of Experiment I, and was e s sent ia l l y a dupl icat ion of that experiment. The fol lowing modifications in methodology were made. 1) Because of a small build-up of mites from greenhouse contamina-t ion of the plants used in Experiment I, the plants for Experiment II and a l l subsequent experiments were seeded d i r e c t l y in the growth chambers under a 12 h photoperiod and 20-23.9°C (68-75°F) temperature regime. This procedure was intended to circumvent the error which would have been introduced in l a te r experiments by the ef fect of changes in daylength and diurnal temperature on the growth of plants in the green-house. 2) The population of wh i te f l i e s from which insects in Experiment I had been drawn was nearly destroyed because of inadequate caging. High humidity ins ide the cage caused a heavy in festat ion of leaf mould, Cladosporium fulvum (Cooke) on the host plants. This s i tuat ion necessitated 31 the use of insects from a chemically contro l led whitef ly population l i v i n g on tomato and eggplant in a d i f fe rent area of the UBC greenhouses. 3) The three cages, representing the three rep l i cat ions on each cu l t i v a r were a l l placed on one rather than two plants of each c u l t i v a r . Plants were seeded on Aug. 8, 1975; the three rep l i cat ions commenced on Sept. 7, 8, and 9, 1975, respect ive ly. Experiment III This experiment was designed to determine whether or not fumigation during the late nymphal instars had any deleterious e f fect on the fecundity and longevity of the emerging adults. I t was thought that such an e f fec t might help to explain some discrepancies between the resu l t s of the f i r s t two experiments. Bush beans (Phaseolus vulgaris L.) were seeded in four 6 - in . p l a s t i c pots on Nov. 19, 1975 and placed in a greenhouse. The select ion of bush beans was made because a fast growing host which would support a s izeable population of wh i te f l i e s was desired. On Dec. 8, 1975, these plants were placed near some heavily infested tomato plants in a section of the greenhouse not subject to fumigation. A l l four pots of bean plants were then kept undisturbed under fluorescent l i gh t s (12-h photoperiod) in a heavily infested condition un t i l Jan. 7, 1976, by which time they supported an abundance of whi tef ly in a l l stages of development. Two of the pots of 32 plants were then moved to a d i f fe rent greenhouse where they were fumigated with "Plantfume 103" (Sulfotep). On the next day (Jan. 8, 1976), these two pots were returned to the o r i g ina l l i ghted greenhouse sect ion. Adult wh i te f l i e s which emerged on Jan. 13, 14, 15, and 16, 1976, from these four pots of bean plants were used on tomato plants in rep l i cat ions I, I I , III and IV of Experiment III respect ive ly. A tota l of eight tomato plants which had been seeded on Dec. 19, 1975 was employed; there were four plants each of cvs. Tiny Tim (hereafter,TT) and Campbell 1327 (hereafter, CB). These cu l t i va r s represented the extremes in glandular hair density and whitef ly s u s cep t i b i l i t y as determined in Experiments I and I I. Replications I and III were placed on four of the plants (two of each c u l t i v a r ) , whereas rep l icat ions II and IV were placed on the other four plants. Each rep l i cate consisted of four leaf cages, two of which contained ten wh i te f l i e s which had received the pre-emergence fumigation treatment. These two cages were placed on e i ther TT or CB. Each of the other two leaf cages per rep l i ca t i on contained ten wh i te f l i e s which had never been fumigated. These were placed on the other two plants, one of TT and CB. Each of the eight plants supported two cages of wh i t e f l i e s , one which had and one which had not received the pre-emergence fumigation. See "Figure 1. The plants and insects were handled as described for the e a r l i e r experiments. 33 Key TT - cv. Tiny Tim CB - cv. Campbell 1327 I - Repl icat ion I II - Repl ication II III - Repl ication III IV - Repl ication IV c - no pre-ernergence fumigation of wh i te f l i e s f - pre-emergence fumigation of wh i te f l i e s Q - cage containing f i ve female and f i v e male wh i te f l i e s ^ - plant Figure 1. Design of Experiment I I I. 34 Experiment IV Several potential sources of more economically valuable resistance to whitef ly were investigated i n consecutive and sometimes simultaneous experiments from Oct., 1975 to Jan. , 1976. The material in which resistance was sought f e l l into three groups: 1) plant material included in the pedigree of cv. Swift 2) csome species of the genus Lycopersicon 3) some woolly mutant stocks. Since the nature of the experiments and the growth chamber environment varied l i t t l e over th i s period, the work has been compiled and the results reported in two units . IPa!rt A: Search for whi tef ly resistance in the breeding background of Swift. Swift was reported by Curry and Pimentel (1972) to possess the highest level of resistance to whitef ly of any cu l t i v a r tested. Its r e l a t i ve resistance was confirmed in the present study. Its background was chosen for study because i t s resistance seemed less var iable than that of other cu l t i var s on the basis of the two procedures for deter-mining res istance, i . e . that of Curry and Pimentel (1972) and that used here.. Figure 2 outl ines the pedigree of Swift with a br ie f descr ipt ion of the va r ie ta l cha rac te r i s t i c s ' o f the plant material employed in th i s study ( a l l those appearing in boxes in Figure 2). 35 Figure 2. Pedigree of Swift and var ieta l descr ipt ion of plant material studied in Experiment IVa. Bison- 4.. p impine l l i fo l ium Farthest North Redskin Swift 1 Bounty F-, se lect ion Cu l t i var Character i st ics L_. pi mpinel l i folium Farthest North Bounty Redskin indeterminate plant, very ear ly y i e l d , small round red f r u i t ( 1 cm. diam.), often used in breeding for earl iness or resistance to diseases, such as Cladosporium fulvum Cooke (Luckwi l l , 1943), determinate p lant, very early y i e l d , smal l , round red f r u i t (Yeager, 1938). determinate plant, ear ly y i e l d , uniform r ipening, medium sized round red f r u i t (Minges, 1972). determinate p lant, early y i e l d , medium sized dark red f r u i t (Yeager, 1938). 36 The procedures for determining glandular hair density, handling of the plants and insects , the source of insects and the seed sowing date were a l l the same as in Experiment II. Oviposit ion rates are based on mean values for the l i fespan of the insects. Part B: Search for whitef ly resistance in the genus Lycopersicon and tomato woolly mutants. The purposes of th i s experiment were to evaluate resistance in two Lycopersicon species [J_. cheesemanii R i ley and U peruvianum M i l l . var. humifusum Mul1. (hereafter, Lph)] which apparently were overlooked by previous investigators (Gentile et al_., 1968; Hogenboom et al_., 1974; de Ponti ejt aj_., 1975), and to examine some plants of L_. hirsutum, a species in which various degrees of whitef ly resistance were reported by a l l of these authors. Morphological descriptions of the Lycopersicon species studied can be found in Luckwil l (1943). Two representatives [Van Wert's woolly (Wov/+)* and Morgan's woolly (Wom/Wom)] of the class of so-ca l led wooljy:tomato mutants were also investigated for whitef ly resistance. Both mutants, which occurred spontaneously in tomato cv. Rutgers 3 (Rick, 1955; Soost, 1957), are characterized by a heavy vesture of nonglandular trichomes over the ent i re aer ia l epidermis. The "wooliness" i s conditioned by various a l l e l e s at a s ingle locus. The woolly genes are dominant to wi ld type and severa l , including the Van Wert woolly, are letha l in the homozygous condition ;(Soost, 1957). A descr ipt ion of the epidermal surfaces of some of the woolly mutants in terms of kinds and density of the various classes of trichomes can be found in Rick and *the plants used were the F, hybrid between Vantage and the Van Wert's woolly mutant. The hybrid was equivalent in wooliness to the o r i g ina l Van Wert's woolly parent. 37 and Butler (1956). The susceptible controls were the greenhouse tomato cvs. Vantage and Vendor. The hybrid [Vantage X J-. hirsutum! was also evaluated. The methodology varied l i t t l e from that of Experiment II. Mean ov ipos i t ion rates are based on data co l lected in the f i r s t 21 days of the experiment. Glandular hair density was determined for Vantage, L_. hirsutum and the i r F-| hybrid. Experiment V The resistance of Lph as determined in Experiment IV was further investigated for e f fect on developmental stages of w h i t e f l i e s . Number of eggs l a i d , development of eggs, mortal i ty during development, and the sex r a t i o of emerging adults were observed on both Lph and the susceptible tomato cv. Tiny Tim (hereafter, TT). Whitef l ies (not f reshly emerged) were col lected randomly from young fo l iage of a _L. p impine l l i fo l ium plant growing in an unfumigated green-house. Ten female and f i ve male wh i te f l i e s were caged on a young l e a f l e t of a plant which was e i ther TT or Lph for 24-25 hours. The wh i te f l i e s and cages were then removed and development of the eggs was observed at two day interva ls un t i l adults were ready to emerge. The l ea f l e t s were recaged at that point and the number and sex of newly emerged wh i te f l i e s were recorded da i l y . During development of the i n s ta r s , the approximate number of wh i te f l i e s in each stage of development was recorded. Five 38 rep l i cat ions were started on consecutive days from Jan. 15 to Jan. 19, 1976. Each rep l i ca t i on commenced on one plant of TT and one of Lph. One l e a f l e t per plant was caged with the 15 wh i t e f l i e s . Experiment VI Two studies using very small whi tef ly populations were conducted to try to gain more understanding of the nature of the resistance of Lph. The f i r s t attempted to determine whether there would be an e f fect on whitef ly performance i f only the f i r s t 24 h a f ter emergence had been spent on the res i s tant plant. The second study compared the fecundity and longevity of wh i te f l i e s which had completed the i r nymphal and pupal development e i ther on the res i s tant Lph plants or on the susceptible tomato cv. TT. Study A: Effect of ear ly feeding on Lph. Freshly emerged wh i t e f l i e s , taken from a TT plant growing in an unfumigated greenhouse, received a pretreatment of 24 h feeding in a growth chamber (Jan. 13-14, 1976) on e i ther four-week old Lph or TT plants. Whitef l ies from each pretreatment group were then caged ( f ive females and f i v e males per cage) on l ea f l e t s of e ither Lph or TT for three days. The wh i te f l i e s and cages were removed and development of eggs through to adult was observed as in Experiment V. Study B: Comparative fecundity and longevity of wh i te f l i e s completing the i r pre-adult development on TT or Lph. Insect or disease resistance in plants i s viewed by many as a f l ee t i ng 39 phenomenon overcome in a few generations by the genetic v a r i a b i l i t y of the pest or pathogen. The longevity and fecundity of wh i te f l i e s completing the i r pre-adult development on Lph i s of considerable interest with respect to th i s view. Whitef l ies emerging da i l y on e i ther Lph or TT in Experiments V and VIA were placed on young, caged l ea f l e t s of e ither Lph or TT un t i l eno had been obtained for the desired number of rep l i ca t i on s . Unfortunately, by this- t ime, (seven to eight days) many of the insects placed on Lph had already d ied, so that only two rep l i cat ions of four cages,each containing four insects (two female and two male), could be set up. The four cages per rep l icat ions contained insects with the fol lowing history: 1) pre-adult development on TT, adults placed on TT 2) pre-adult development on TT, adults placed on Lph 3) pre-adult development on Lph, adults placed on TT 4) pre-adult developmentvon Lph, adults placed on Lph Oviposit ion rate and longevity were determined as i n Exp. IV. 40 RESULTS Experiment 1. Var ieta l e f fect on whitef ly character i s t i c s in some cu l t i va r s of L_. esculentum. The tomato c u l t i v a r on which adult wh i te f l i e s were reared had a s i gn i f i can t e f fect (P < 0.05) on the longevity, ov ipos i t ion rate and fecundity of the insects (Table IV). Longevity ranged from a mean of 19.0 days on Campbell 1327 to 47.4 days on Roma VF. Whitef l ies l i ved s i g n i f i c an t l y longer on Roma VF and Tiny Tim than on the other three cu l t i v a r s . There were no s i gn i f i c an t differences in longevity among rep l icat ions or between sexes. Oviposit ion rate ranged from a mean of 3.9 eggs per female per day on Swift to 8.4 eggs per female per day on Tiny Tim; the rate on Tiny Tim was s i g n i f i c an t l y greater than those fo r wh i te f l i e s on the other four cu l t i v a r s . Fecundity of wh i te f l i e s ranged from a mean of 73.3 eggs per female on Campbell 1327 to 372.1 eggs per female on Tiny Tim. S i gn i f i can t l y more eggs were deposited on Tiny Tim than on the other four cu l t i va r s and on Roma VF than on three other cu l t i v a r s . [See Appendix I for ANOVA tab les . ] The density of glandular hairs on the upper leaf epidermis was highly var iab le, even on leaves of s im i la r age from the same plant, but there were s i gn i f i can t differences in mean densit ies among the f i v e c u l t -ivars (Table IV). The mean density on 4 mm of the upper leaf epidermis ranged from 29.0 hairs on Tiny Tim to 102.7 on Campbell 1327. The l a t t e r value was s i g n i f i c an t l y greater than those for the other four c u l t i v a r s ; means for Tiny Tim and Roma VF were s i g n i f i c an t l y lower than those for the other three cu l t i v a r s . There were no s i gn i f i can t differences in 41 Table IV. The e f fect of tomato cu l t i v a r on means for whitef ly longevity, ov ipos i t ion rate and fecundity, and glandular hair density on both upper and lower leaf epidermis of the cu l t i va r s (Experiment I ) . * Tomato Cu l t i var Variable Tiny Tim Roma VF Swift Sunray Campbell 1327 Whitefly Longevity* days sample s ize 40.4a (29) 47.Sab (28) 29.1bc 27.5bc (28) (28) 19.0c (28) Daily ov ipos i t ion rate eggs/? 8.4a 5.7b 3.9b 4.4b 4.0b Fecundity tota l eggs/? 372.1a 272.5b 116.0c 126.5c 73.3c Var ietal GHD-UE no. of hairs/4 mm2 29.0a 43.7a 67.1b 81.4b 102.7c GHD-LE no. of hairs/4 mm2 31.3a 26.2a 24.2a 30.8a 30.5a *rneans include both sexes. tmean separation within rows by Student-Newman-Keuls mult ip le range te s t , 5% l e v e l . 42 glandular hair density on the lower leaf surfaces of the f i v e c u l t i v a r s . S ign i f i cant negative corre lat ion coe f f i c i en t s were obtained between mean glandular hair densit ies on the upper leaf epidermal surfaces and longevity (r = -0.89, P < 0.05) and fecundity (r = -0.94, P < 0.05) of wh i te f l ie s on the f i v e cu l t i v a r s . The cor re la t ion coef f i c ient s between mean glandular hair density on the upper leaf surfaces of the f i v e cu l t i va r s and ov ipos i t ion rate (r = -0.79, P < 0.11) and percentage of eggs deposited on that surface (r = -0.69, P < 0.20) were not s i g n i f i c an t , although the former approached s ign i f i cance. On each c u l t i v a r , whitef ly ov ipos i t ion rate i n i t i a l l y increased with age of the insects (Figure 2). Oviposition rate declined gradually a f ter atta in ing a maximum value at a mean insect age of from 12 to 20 days, depending on the tomato cu l t i v a r on which the wh i te f l i e s were reared. Occasional f luctuat ions in th i s decline can be observed at points where fresh plant material was introduced. Because var ieta l influence on whitef ly ov ipos i t ion rate can be detected within the f i r s t three weeks of the experiment (Figure 3), data for ov ipos i t ion rate • - only co l lected v ^ J t t f f \ ^ IA- Cr for 21 days in l a te r experiments. y Figure 3. Oviposition rate vs. age of female wh i te f l i e s as observed on f i ve tomato cu l t ivars in one rep l i ca t i on of Experiment I. Numbers along the curves represent female insects remaining in each cage. Tiny Tim Roma VF Sunfay - Swift Campbell 1327 t Change to new plants J 1 1 1 l i L 40 + . 50 60 70+ 80 ..Age of whitef 1 ies. (days)•• 44 Experiment II. Duplication of Experiment I with methodological modif icat ions. Although the changes in technique of Experiment II did not much a l t e r the impl icat ion of the resu lts of Experiment I, t he i r s t a t i s t i c a l s ign i f i cance was generally lower in Experiment I I. Longevity ranged from a mean of 16.5 days on Campbell 1327 to 41.9 days on Roma VF (Table V). Whitefl ies again l i ved s i g n i f i c an t l y longer on Roma VF than on the other four c u l t i v a r s , but mean longevity on a l l f i v e cu l t i va r s was less than i t had been in Exeriment I. There was no s i gn i f i c an t di f ference in mean longevity among rep l i cat ions or between sexes. The s i g n i f i c an t l y longer l i fespan observed on Tiny Tim in Experiment I was not observed in Experiment II. Oviposit ion rate and fecundity were again lowest on Campbell 1327„-(4.3 eggs per female per day, 96.6 tota l eggs per female) and highest on Tiny Tim (6.6, 195.8), but neither t r a i t varied s i g n i f i c a n t l y among cu l t i va r s (see Appendix I I ) . In spite of rank order differences among the f i ve cu l t i va r s with respect to whitef ly longevity and da i l y ov ipos i t ion rate between the two experiments, the rank order among cu l t i va r s with respect to to ta l fecundity remained unchanged. The f i ve cu l t i va r s examined in Experiment II were ranked according to density of glandular hairs on the upper lea f epidermis in approximately the same order as in Experiment I (Table V). Tiny Tim and Campbell 1327 were again found to have the most sparse and most dense vesture of glandular hairs respectively on the upper epidermis, but the mean density was lower than that measured in Experiment I fo r each of these c u l t i v a r s . On the upper leaf epidermis, Tiny Tim had s i g n i f i c an t l y fewer glandular hairs 45 Table V. The e f f e c t of tomato cu l t i v a r on means for whitef ly longevity, ov ipos i t ion rate and fecundity, and glandular hair density on both upper and lower leaf epidermis of the cu l t i var s (Experiment I I ) . Tomato Cu l t i var Variable • Tiny Tim Roma VF Swift Sunray Campbell 1327 Whitefly Longevity* days sample s i ze 27.8at (26) 41.9b (28) 28.2a (26) 24.2a (15) 16.5a (27) Daily ov ipos i t ion rate eggs/? 6.6 4.6 4.7 5.4 4.3 Fecundity tota l eggs/? 195.8 183.9 128.1 154.7 96.6 Var ieta l GHD-UE no. of hairs/4 mm2 24.0a 48.9b 65.1c 40.5b 75.0c GHD-LE no. of hairs/4 mm2 19.9 24.3 21.4 19.6 25.0 *rneans include both sexes. tmeanr- separation within rows by Student-Newman-KeuIs mult ip le range tes t , 5% l e v e l . 46 than the other four c u l t i v a r s , whereas Swift and Campbell 1327 had s i g n i f i c an t l y more glandular hairs than the other three c u l t i v a r s . The Sunray plants examined in Experiment II were found to have a much lower mean glandular hair density on the upper lea f epidermis than those examined in Experiment I; a corresponding increase in whitef ly ov ipos i t ion rate on th i s c u l t i v a r was also observed. There were no s i gn i f i can t differences in glandular hair densit ies on the lower lea f epidermis of the f i v e c u l t i v a r s . S i gn i f i cant negative cor re lat ion coe f f i c ient s were found between glandular ha i r density on the upper leaf epidermis and 1) ov ipos i t ion rate (r = -0.91, P < 0.05) and 2) fecundity (r = -0.90, P < 0.05) of wh i te f l i e s on the f i ve c u l t i v a r s . The cor re lat ion coe f f i c i en t between density of glandular hairs on the upper leaf epidermis and longevity (r = -0.35, P < 0.50) of wh i te f l i e s on the f i v e cu l t i va r s however, was not s i gn i f i c an t . 47 Experiment I I I. Ef fect of pre-emergence fumigation on character i s t i c s of adult wh i t e f l i e s . Fumigation during the late nymphal and pupal ins tar stages appeared to have had no e f fec t on the longevity or fecundity of the adults emerging within a few days (Table VI). Mean ov ipos i t ion rates, varied l i t t l e over the four treatments; the differences were not s i gn i f i can t . Whitefly longevity varied with the cu l t i v a r as in 'the two previous experiments. Regardless of fumigation or sex of the insects, the mean l i fespan of wh i te f l i e s reared on Tiny Tim was s i g n i f i c an t l y longer (26.7 days) than on Campbell 1327 (15.7 days) (Appendix I I I ) . Since there was l i t t l e d i f -ference in ov ipos i t ion rate over the four treatments, fecundity would merely r e f l e c t the differences in longevity; analysis of fecundity therefore has been excluded. From the data in Table VI, no consistent e f fect can be attr ibuted to the fumigation treatment. Mean longevity on either cu l t i v a r i s nearly ident ica l to that of Experiment I I, but considerably less than that of Experiment I. This suggests that the plant handling technique which was consistent in Experiments II and I I I , but modified since Experiment I, may account for the differences observed between Experiments I and I I, or I and I I I . If the ov ipos i t ion rates obtained in Experiment III are analyzed only with respect to the plant on which the insects were caged rather than fumigation treatment and c u l t i v a r (Table VI I ) , then s i gn i f i can t p lant-to-plant differences can be found (see Appendix I I I ) . Both the plant population s ize and the number of observations per plant are l i m i t i n g , but 48 Table VI. Means for longevity and ov ipos i t ion rate of wh i te f l i e s as affected by pre-emergence fumigation ( ' Su l fotep ' ) vs. control on two tomato cu l t i var s (Experiment I I I ) . Tomato Cu l t i var whitef ly t r a i t Tiny Tim Campbel1 1327 control fumigated control fumigated Longevity* days sample s ize 24.8at (37) 28.6a (39) 15.4b (40) 16.0b (40) Daily ov ipos i t ion rate eggs/? 5.2 5.5 5.7 5.5 *means include both sexes. tmean separation within the row by Student-Newman-Keuls mult ip le range tes t , 5% l e v e l . Table VII. Mean ovipos i t ion rates obtained in Experiment III arranged according to the plant on which they were recorded. Plant Fumigation treatment Repl icat ion Oviposit ion rates Tiny Tim Campbell 1327 1 + I III 5.77 4.50 7.58 7.28 2 + I III 5.92 6.41 4.53 5.38 3 + II IV 4.74 5.01 3.56 5.50 4 + II IV 5.38 5.24 5.29 5.36 +pre-emergence fumigation with Sulfotep -control 49 the data indicate that the influence on ov ipos i t ion rate of an indiv idual plant on which a whitef ly population feeds i s greater than var ie ta l or pre-emergence fumigation e f fec t s . Stoner et al_. (1968.and 1969) have also found s u f f i c i en t var iat ion in resistance to aphid (Macrosiphum euphorbiae) and f r u i t f l y (Drosophila melanogaster) in plants of the same tomato cu l t i v a r or breeding l i ne to recommend se lect ion within the l i ne or c u l t i v a r . 50 Experiment IVa. Search for resistance to whitef ly in the pedigree of Swift. Density of glandular hairs on e i ther leaf epidermis and wh i te f ly fecundity and longevity varied considerably among L_. p impine l l i fo l ium and three tomato c u l t i v a r s , a l l of which had been employed in the breeding program for the development of Swift ; however, no plant of these cu l t i va r s or accession appeared to be more res i s tant to whitef ly than Swift i t s e l f . Mean whitef ly longevity ranged from 19.9 days on Redskihrto 42.8 days on Farthest North. Mean l i fespan was s i g n i f i c an t l y longer on Farthest North-.than on the other three cu l t i va r s and L_. p impine l l i fo l ium and s i g n i f i c an t l y shorter on Redskin and Swift than on the other two cu l t i va r s and L. p impine l l i fo l ium. There was no s i gn i f i c an t difference in longevity between rep l i cat ions or sexes (see Appendix IV). Mean fecundity ranged from 70.4 eggs per female on Swift to 299.7 eggs per female on Farthest North. Mean values for ov ipos i t ion rate ranged from 3.9 eggs per female per day on Swift to 6.7? eggs per female per day on Farthest North; however, the differences in both fecundity and ov ipos i t ion rate were not s t a t i s t i c a l l y s i gn i f i can t (Table V I I I ) . It was expected that cu l t i va r s with a high density of glandular hairs and consequently greater resistance to whitef ly would be found in the pedigree of Swift . Density on e i ther leaf epidermis was shown to d i f f e r s i g n i f i c an t l y among L_. p impine l l i fo l ium and the four cu l t i va r s 2 in th i s pedigree. The mean number of hairs per 4 mm on the upper leaf surface ranged from 17.7 on j _ . p impine l l i fo l ium to 73.5 on Swift ; I. pimpine l l i fo l ium and Farthest North had s i g n i f i c an t l y fewer glandular 51 -Table VII I. The e f fect of L_. p impine l l i fo l ium and three tomato c u l t i v a r s , a l l of which appear in the pedigree of "Swift , compared with the e f fect of Swift on means for longevity, ov ipos i t ion rate and fecundity of wh i te f l i e s and means for glandular hair density on e i ther f o l i a r epidermis of the plant material (Experiment IVa). Tomato Cu l t i var Variable L. uimpinel- Farthest Bounty Redskin Swift l i f o l i u m North Whitefly Longevity* days sample s i ze Daily ov ipos i t ion rate eggs/? Fecundity tota l eggs/$ Var ieta l GHD-UE no. of hairs/4 mm2 GHD-LE no. of hairs/4 mm2 32.2 (20) 42.8 32.3 19.9 20.9 (20) (20) (20) (15) 5.6 6.7 4.5 4.6 3.9 156.3at 299.7b 131.4a 79.3a 70.4a 17.7a 19.9a 57.8b 60.4b 73.5c 15.9b 6.9c 22.6a 16.7b 12.2bc *means include both sexes. tmean separation within rows by Student-Newman-Keuls mult iple range t e s t , 5% l e v e l . 52 hairs than the other three cu l t i va r s whereas Swift had s i g n i f i c a n t l y more hairs on that surface than 4. p impine l l i fo l ium or the other three c u l t i v a r s . The mean value obtained for Swift (73.5 hairs per 4 mm ) i s comparable to o those obtained in Experiments I and II (67.1 and 65.1 hairs per 4 mm respect ive ly ) , lending support to the hypothesis that glandular hair density on Swift i s not as var iable as has been shown in other c u l t i v a r s . 2 On the lower leaf epidermis, glandular hair density per 4 mm ranged from 6.9 on Farthest North to 22.6 on Bounty; the values for Farthest North and Swift were s i g n i f i c an t l y smaller and that for Bounty was s i g n i f i c an t l y greater than those for L_. p impine l l i fo l ium arid other cu l t i va r s studied. Rank order for glandular hair densit ies on the lower epidermis i s remarkably d i f fe rent from rank order on the upper epidermis. S i gn i f i cant negative cor re lat ion coe f f i c ient s were obtained between mean glandular hair density on the upper leaf epidermis of the f i v e c u l t i -vars and mean ov ipos i t ion rate (r = -0.93, P < 0.05) and mean fecundity (r = -0.90, P < 0.05) of wh i t e f l i e s . The negative cor re la t ion coe f f i c i en t between means for hair density on the upper leaf epidermis and longevity (r = -0.80, P < 0.10) was nearly s i gn i f i c an t . Mean glandular hair density on the lower leaf epidermis and the three whitef ly character i s t i c s were negatively but not s i g n i f i c an t l y correlated [ov ipos i t ion rate (r = -0.57, P < 0.35), longevity (r = -0.34, P < 0.50), fecundity (f = -0.55, P < 0.30)]. 53 Experiment IVb. Search for resistance to wh i te f l i e s in the genus Lycopersicon and the tomato woolly mutants. Extremes of both s u s cep t i b i l i t y and resistance to whitef ly were revealed in the d i ver s i t y of plant material examined in th is experiment. J_. peruvianum var. humifusum (Lph) and Van Wert's woolly mutant were i den t i f i ed as potential sources of resistance when the longevity, ov ipos i t ion rate and fecundity of wh i te f l i e s reared on these plants were shown to be consistently lower than on the other plant material (Table IX). The plant mater ia l , the sex of the insects and the interact ion term between plant material and sex were a l l s i gn i f i can t sources of variance fo r longevity of wh i te f l i e s (see Appendix IV). With the exception of insects reared on Lph, females on average l i ved longer than males or else there was no difference in mean l i fespan between the sexes. Females on Lph l i v ed approximately half as long as males; only s ix insects , a l l male, out of 57 managed to survive longer than two weeksron plants of th is species. There was a very s i gn i f i can t difference in mean l i fespan for both sexes observed on the two woolly mutants; th i s l i k e l y ref lected the difference in trichomal morphology between these mutants'. The range of ov ipos i t ion rates (0.35 to 12.80 eggs per female per day on Wov/+ and _L. cheesemanii respect ively) was much greater than has been observed in previous experiments. More eggs were deposited da i l y on j . . cheesemanii than on any plant material examined in th is study or reported in the l i t e r a t u r e . There was also a s i gn i f i can t difference in mean ov ipos i t ion rates on the two greenhouse cu l t i va r s studied. The rate on Vantage (3.47 eggs per female per day) was less than half that on Table IX. Ef fect of some Lycopersicon accessions, tomato woolly mutants and greenhouse tomato " cu l t i v a r s on means for longevity, ov ipos i t ion rate and fecundity of whitefl ies.(Experiment IVb). Whitefly t r a i t Greenhouse cu l t i va r s Vendor Vantage Lycopersicon accessions F, [Van- L_. h i r su- U cheese-tage x L. turn manii hirsutum] Lph Woolly mutants Wom/Wom Wov/+ Longevity dadays ? o" sample size(+/(?) ov ipos i t ion rate 7.66e eggs/?/day Fecundity to ta l eggs/? 27.2ct 18.9b (13/13) 3.47bc 91.7b 17.5b 17.3b 6.60de 129.4b 23.3bc 19.0b (15/15) (14/14) * * 4.08bcd 12.80f 89.3b 4.3a 1:10?2a (30/27) 2.02ab 8.8a 42.5d 29.3c 5.8a 5.8a (13/11) (15/15) 5.35cde 0.35a 225.9c 2.7a *these values were not obtained because of a growth chamber malfunction which caused the premature death of a l l remaining insects at 22 days. tmean separation within rows by Student-Newman-Keuls multiple range te s t , 5% l e v e l . 55 TABLE X. Mean density of glandular hairs on Vantage , F, (Vantage x L. hirsutum) and L. hirsutum. Glandular hai r density Vantage Plant Material F-, [Vantage x L_. hirsutum] L. hirsutum Upper Epidermis Lower Epidermis 53.22at 16.36a 39.86b 26.43b 61.93a 49.89c tmean separation within rows by Student-Newman-Keuls mult iple range te s t , 5% l e v e l . Vendor (7.66 eggs per female per day). Fecundity ranged from 2.7 eggs per female on Wov/+ to 225.9 eggs per female on Wom/Wom; s i g n i f i c an t l y more eggs were deposited;'per female on Wom/Wom whereas s i g n i f i c an t l y fewer eggs were deposited per female on Wov/+ and Lph than on the other plant material studied. Glandular hair densit ies were measured for Vantage, L_. hirsutum and the i r F^ hybrid (Table X). Density on the upper leaf epidermis was s i g n i f i c an t l y lower on the F-| hybrid than that on e ither parent. On the lower leaf epidermis, Vantage had s i gn i f i c an t l y fewer glandular hairs while L_. hirsutum had s i g n i f i c an t l y more glandular hairs than the F-| hybrid. This anomolous expression of glandular ha i r density indicates a complex pattern of inheritance. Since glandular hairs are very sparsely d i s t r ibuted on Lph and nonexistant on Van Wert's woolly mutant, both sources of resistance suggest mechanisms other than glandular ha i rs . 56 Experiment V. Comparative development of whitef ly on res i s tant and susceptible host plants. Experiment V (Table XI) indicates that wh i te f l i e s can complete the i r development on the res i s tant plants of J_. peruvianum var. humifusum (Lph), but that mortal i ty during development is s i g n i f i c an t l y greater than that on the susceptible tomato cv., Tiny Tim (TT). Morta l i ty ranged from 0 to 48% on TT (mean = 27%) and from 53 to 81% on Lp_h , (mean = 67%). S i gn i f i can t l y more adults emerged on TT than on Lph, and there was also s i gn i f i can t var iat ion among rep l icat ions with respect to the number of adults emerging. These effects may have been due to p lant-to-plant var iat ion in resistance because each rep l i ca t i on was located on a d i f fe rent pair of plants. Observations during development indicated that most of the mortal i ty on boith hosts occurred during the th i rd and fourth in s ta r s . The actual number of eggs l a i d on each leaf was not counted in order to avoid damage to the leaves or dislodgement of the eggs. Since the number of eggs which hatched did not d i f f e r s i g n i f i c an t l y on e i ther res i s tant or susceptible p lants, i t can be assumed that ov ipos i t ion rate was the same on Lph and TT. The sex ra t io of the progeny appeared to favor males in a l l but two cages, both occurring? on Lph. There were no s i gn i f i c an t rep l i ca t i on or var ie ta l e f f ec t s , however, on the proportion of males. Both male and female wh i te f l i e s required s i g n i f i c an t l y more time (approx-imately one day) to develop on Lph than on TT. Mean development times for both sexes were ident ica l on Lph whereas males required an addit ional day more than females to develop on TT. The interact ion term (sex X var iety) in the analysis of variance was approaching s ign i f icance (Appendix V). Data for development time in each sex on the two species are presented graphical ly in Figure 4. 57 TabletXI. Comparative development of whi tef ly on the susceptible tomato cv. Tiny Tim (TT), and the res i s tant accession of L_. peruvianum var. humifusum (Lph) (Experiment V). No. of eggs No. of Morta l i ty sex mean Replication Host hatched: adults during ra t i o development emerged: development of time •{%): adults (days) Mean TT 36 30 17 10/20 27.2/28.0 Lph 38 18 53 13/5 29.1/27.8 TT 33 17 48 4/13 26.8/28.3 Lph 32 6 81 : l/5 27.0/29.0 TT 34 22 35 9/13 26.8/29.5 Lph 14 6 57 1/5 29.0/29.0 TT 32 32 0 11/21 27.1/27.6 Lph 30 8 73 2/6 30.0/31.0 TT 33 21 36 8/13 27.0/28.5 Lph 30 9 70 6/3 29.5/29.0 TT 33.6at 24.4a 27a 42/80 27.0/28.3a Lph 28.8a 9.4b 67b 23/24 29.2/29.2b tmean separation within columns by Student-Newman-Keuls mult ip le range te s t , 5% l e v e l . 58 12 24 25 26 27 28 29 30 31 32 33 34 35 days a f te r egg deposition 59 Experiment Via. Effects on whitef ly t r a i t s of early adult feeding on Lph. Feeding by wh i te f l i e s on _L. peruvianum var. humifusum (Lph) during the f i r s t 24 h in the l i f e of the imago appeared to influence the ov ipos i -t ion rate over the next three-day period, mortal i ty during development and the sex ra t i o of emerging adults (Table XI I ). No eggs were l a i d by female wh i te f l i e s during the f i r s t 24-h adult feeding treatment. Although there were no rep l i ca t i on s , inspection of the data in Table XII shows the fol lowing trends: 1) Whitef l ies which were caged on Lph for the f i r s t 24 h of adult l i f e deposited more eggs on each host respect ively than the i r counterparts which were caged on TT for the i n i t i a l 24 hours. 2) A greater mortal i ty during development occurred (a) i f the parents spent the i n i t i a l period on TT and were transferred to Lph (81%) than i f they were reared continuously on Lph (53%); (b) i f the parents spent the i n i t i a l period on Lph and were transferred to TT (42%) than i f they were reared continuously on TT (9%) or (c) i f development occurred on Lrjh_ (81% and 53%) rather than TT (9% and 42%). The t rans fer ra l i t s e l f was not responsible for these differences because even whi tef l ie s reared continuously on the same host were transferred to another l e a f l e t a f ter the i n i t i a l 24 hours. 3) If the parents spent any time on Lph, the sex r a t i o of the progeny was heavily skewed in favor of males. This is pa r t i cu l a r l y Table XII. Comparative development of wh i te f l i e s whose parents spent varying amounts of time on res i s tant (Lph) and susceptible (TT) host plants (Experiment V ia) . Treatment group*: F i r s t 24 h of adult l i f e on: Next 24-96 h of adult l i f e on: No. of 1st nymphal instars hatched: No. of adults emerged: Mortal i ty during de-velopment (%): sex r a t i o of adults I II III IV TT TT Lph Lph TT Lph TT L-gh 56 26 89 .32 51 5 52 15 9 81 42 53 25/26 1/4 4/48 4/11 *Nymphal development for a l l groups occurred on TT. 61 apparent in treatment group I I I , in which adults were caged for 24 h on Lph and were transferred to TT for the three days when egg deposition occurred. Experiment VIb. Effect on t r a i t s of adult wh i te f l i e s of pre-emergence development on Lph. Pre-emergence development of whitef ly on res i s tant Lph plants appeared to a f fec t ov ipos i t ion rate and fecundity de leter ious ly regardless of whether adult feeding occurred on the susceptible TT or the res i s tant Lpji plants (Table XI I I ). Although the population of wh i te f l i e s was smal l , inspection of data in Table XIII suggests the fol lowing trends. 1) Although reproductive success of wh i te f l i e s i s reduced by development on Lph, female wh i te f l i e s that have completed development on Lph are s u f f i c i e n t l y fecund to deposit large numbers of eggs on a susceptible host plant. 2) Adults which developed on Lph are no more to lerant of the res i s tant host than adults which developed on TT. 3) Feeding by adults has a much greater influence than nymphal feeding on the fecundity of adults. 62 Table XI I I. Influence of susceptible (TT) and res i s tant (Lph) host plants on character i s t i c s of whitef ly adults whose pre-emergence development occurred on e i ther TT or Lph (Experiment VIb). Whitefly Development on: Adult feeding°on: Host Plant Lph Lph Lph TT TT Lph —i —i Longevity* days 10.5 26.5 12.5 26.7 sample s i ze (8) (10) (8) (9) Oviposit ion rate eggs/? 2.2 6.7 3.1 8.3 Fecundity tota l eggs/? 27.0 177.6 43.4 263.1 *means include both sexes. 63 DISCUSSION In the l i t e r a t u r e on insect resistance in Lycopersicon, thetidensity of glandular hairs has often been c i ted as a mechanism of res istance. Curry and Pimentel (1971 and 1972) o f fer no suggestion concerning the factor(s ) underlying the differences they observed in whitef ly fecundity, longevity or population s ize on various tomato c u l t i v a r s . In the planning phase of th i s study, i t was hoped that a pa ra l l e l study of glandular hair density and whitef ly character i s t i c s on some of the same tomato cu l t i va r s studied by Curry and Pimentel (1972) might reveal a cor re lat ion between leaf morphology and estimates of whitef ly parameters, and thus contribute to an understanding of the i r re su l t s . Indeed, s i g n i f i c an t negative correlat ions between mean glandular hair density on the tomato cu l t i va r s and whitef ly ov ipos i t ion rate, longevity and fecundity were found repeatedly in Experiments I, II and IVa. Observations on the glandular hairs as a mechanism of resistance to wh i te f l i e s are consistent with those of Gibson (1971), working with aphids (Myzus persicae, Macrosiphum euphorbiae). Whitef l ies in the present study were occasionally found struggling to free themselves from the s ta lk of a glandular ha i r . In add i t ion, older wh i te f l i e s that had been caged on tomato cu l t i va r s with high densit ies of glandular hairs were seen to have accumulated considerable amounts of blackened exudate on the i r appendages. 64 Although wh i te f l i e s spend most of the i r l i ve s on the lower leaf epidermis, the glandular hair density on the upper lea f epidermis appears to have a profound ef fect on them. One possible explanation i s that the glandular hairs release a tox ic v o l a t i l e o i l in addit ion to or in com-bination with the exudate. Another i s that the exudate accumulates more rapidly on the insects bodies when they have been disturbed, because in such instances, wh i te f l i e s tend to f l y about without d i rect ion and often a l i gh t on the upper leaf surface. The fact that no cor re lat ion was found between glandular hair density on the upper leaf surfaces of several tomato cu l t i va r s and percentage of whitef ly eggs deposited on that surface suggests that the presence of few or many glandular hairs i s not a major factor in a female wh i te f l y ' s se lect ion of an ov ipos i t ion s i t e . Observations made in the present study indicated that wh i te f l i e s do not expend any e f f o r t to avoid contact with glandular ha i r s . This was deduced by observing under a dissecting microscope (lOx) the movement of wh i te f l i e s on leaf surfaces. The insects did not hesitate when confronted by leaf veins where the density of glandular hairs i s t y p i c a l l y great, pa r t i c u l a r l y on the lower epidermis. I t seems possible that wh i te f l i e s are completely obl iv ious to the re lat ionsh ip between glandular hairs and the i r incapacitat ing exudate. The density of glandular hairs on the tomato cu l t i va r s studied in th i s research was not great enough for economically valuable control of wh i t e f l i e s . The observed differences in longevity and fecundity of wh i te f l i e s on the various cu l t i va r s would merely influence the speed 65 with which an unmanageable whitef ly population would develop. In th i s l i g h t , i t i s noteworthy that the changeover in recent years by commercial greenhouse operators in B r i t i s h Columbia from the tomato cv. Vantage to Vendor has probably aggravated the loca l whitef ly problem. The results of Experiment IVb have shown that wh i te f l i e s increase approximately twice as rapidly on Vendor as on Vantage. From the work of Plage (1975), Clayberg and Kring (1974), Clayberg (1975), and de Ponti eJL al_« (1975), i t seems probable that glandular hair density can be increased in tomato c u l t i v a r s , perhaps by intergeneric hybr id izat ion with Solanum p e n n e l l i i , to a level which w i l l provide some degree of contro l . However, the increased opportunity for transmission of sys-tems es v i ruses ' ins.such'a cult-i var , ' as -Harr i s and Bradley (1.973) have warned, might negate the benefits of whitef ly con t ro l , espec ia l ly i f th i s control i s incomplete. I f control were complete or nearly so, however, then i t i s l i k e l y that virus transmission by wh i te f l i e s would be cont ro l led. In addit ion to the s i gn i f i can t negative corre lat ions found between glandular ha i r density and whitef ly cha rac te r i s t i c s , another unquantified observation (with respect to the plant) was made: a high density of nonglandular trichomes appeared to be correlated with a long l i fespan in the wh i t e f l i e s . The highest densit ies of unbranched, nonglandular trichomes were observed on Roma VF and Morgan's woolly mutant, and each of these was correlated with a high mean longevity value of 47.4 (Exp. I) and 36.5 (Exp. IVb) days respect ive ly. Roma VF also had a moderately 66 dense vesture of glandular trichomes; however, longevity, but not fecundity, was s i g n i f i c a n t l y greater on Roma VF than on Tiny Tim, a cu l t i v a r char-acter ized by considerably lower densit ies of both sorts of trichomes. I t was noticed that the eggs tended to be deposited, often in small batches, d i r ec t l y at the base of these simple trichomes. Perhaps there i s some benef ic ia l e f fec t for the insects, e.g. a kind of cleansing act ion, from contact with simple nonglandular trichomes. The dense branched trichomes (Plate 2) of Van Wert's woolly mutant appeared to present a simple mechanical bar r ier to whi tef ly feeding and ov ipos i t ion . This source of resistance was not further invest igated, however, because the resistance mechanism i t s e l f seemed to be at the expense of plant product iv i ty (compare leaf anatomy with that of Vendor, Plate 3). The dense vesture seemed to inter fere s u f f i c i e n t l y with photosynthesis to retard the plant considerably. Moreover, the f r u i t qua l i ty was poor, perhaps due to p le io t rop ic ef fects of the woolly a l l e l e ; even the epidermis of the f r u i t s had dense branched trichomes, a condition which would ce r ta in l y reduce the i r market value. Shil l ing„(1973) has found that tota l protein and amino acid synthesis are decreased in plants carrying the woolly a l l e l e (Wo/+). Oviposit ion rates often varied markedly even among rep l i cat ions on the same c u l t i v a r . Because of the multitude of factors which a f fect whitef ly ov ipos i t i on , i t i s an extremely d i f f i c u l t charac te r i s t i c to estimate accurately. In the present study, small populations rather than 67 Plate 2 . Portion of upper epidermal surface of a young l e a f l e t of Van Wert's woolly mutant. Note recurved leaf margins and absence of glandular hairs (approx. 5X) . 6 3 Plate 3 . Portion of upper epidermal surface of a young l e a f l e t of tomato cv. Vendor, Small white specks are the 4 - c e l l e d t ips of glandular hairs. Simple unbranched trichomes can be seen on the lower portion of the leaf (approx. 5 X ) . 69 indiv idual insects were employed since these ref lected more accurately the conditions of a natural in festat ion and reduced the handling time and number of cages necessary. This procedure resulted in one mean ov ipos i t ion rate value for every f i ve female wh i t e f l i e s . I f one of the f i ve females was pa r t i cu l a r l y unproductive, a f a i r l y common occurrence according to Burnett (1949), then the mean value would have been biased. In add i t ion, since ov ipos i t ion rate has been shown to be a function of age of female wh i te f l ie s in Experiment I, one reason for variance among mean ov ipos i t ion rates on the same cu l t i v a r can be attr ibuted to the vary-ing l i fespans of female wh i te f l i e s from one rep l i ca t i on to another. Another unexpected influence on ov ipos i t ion rate was the replacement of eight-week o ld 'p lants with four-week old plants. Because most tomato cu l t i va r s bloomed between f i ve and s ix weeks under the growth chamber condit ions, the major difference between four-week old and eight-week old plants was an absence of f r u i t or flowers at four weeks, but an abundance of each by eight weeks. I t seems plaus ible that leaves of the older plants were less nutr i t ious fo r the insects because more of the p lants ' reserves were being channeled into reproduction. Genetic var iat ion in the whitef ly population and environmental influence on i t could also have contributed to experimental er ror . During the summer months, outdoor populations of whitef ly occur sporadical ly over the Lower Mainland, especia l ly in the v i c i n i t y of commercial green-houses. There i s plenty of opportunity for wind dispersal of the. 70 populations with subsequent interbreeding among them. Garman and Jewett (1922) have stated that wind i s the chief agent for whitef ly d i spersa l . During the winter months, however, much inbreeding occurs within a re s t r i c ted greenhouse population, and the e f fect of such inbreeding on the insects i s unknown. The effects of seasonal changes in photoperiod, diurnal temp-erature f l uc tua t ion , humidity, or l i g h t in tens i ty on whitef ly fecundity or longevity are also matters fo r speculat ion; ce r ta in l y , there are marked effects on the host plant, and i t s nut r i t i ona l qua l i ty for the developing insects. Maintenance of the insect population under a r t i f i c i a l conditions might have eliminated some of these var iab les , but might have simultaneously created a laboratory animal that was not f u l l y representative of the greenhouse pest. Although the resu lts of Experiments I and II were s im i l a r , there was a greater y i e l d of s t a t i s t i c a l l y s i gn i f i can t comparisons in Experiment I. The maximum values for longevity on Tiny Tim and Roma VF, ov ipos i t ion rate on Tiny Tim and glandular hair density on the upper leaf epidermis of Campbell 1327 were a l l markedly lower in Experiment II (Tables IV and V). Considering the procedural differences between Experiment I and Experiment II, the two populations (unfumigated vs. fumigated) from which wh i te f l i e s were drawn seems a less l i k e l y cause of var iat ion than the d i f fe rent plant handling techniques (greenhouse vs. growth chamber grown for the f i r s t 30 days). It i s rea l i zed that the single fumigation of the whitef ly population employed in Experiment III could hot have 71 duplicated the regular periodic fumigation with a sequence of chemicals to which the whitef ly population in Experiment II had been subjected. However, the results of Experiment III suggest that fumigation during the la te instars does not appreciably a f fect fecundity or longevity in the developing insects. The greenhouse-grown plants (June-July, 1975) of Experiment I had a longer photoperiod, higher l i g h t in tens i ty and per iod ica l l y warmer temperatures than the growth chamber-grown plants of Experiment II. These conditions produced very vigorous plants with more succulent leaves, having a greater density of glandular hairs and per-haps a better i n i t i a l plane of nu t r i t i on for the wh i t e f l i e s . Such plants might be expected to have influenced the greater fecundity and longevity of the wh i t e f l i e s , pa r t i cu l a r l y on cu l t i va r s s t i l l r e l a t i v e l y sparsely covered with glandular hairs . The levels of fecundity of wh i te f l i e s caged on the f i v e tomato cu l t i va r s studied in Experiments I and II bear l i t t l e resemblance to the " res istance" of the same cu l t i va r s as reported by Curry and Pimentel (1972) (Table XIV). The posit ion of cv. Campbell 1327 i s of pa r t i cu la r i n te res t ; in the present study, i t appears most " r e s i s t an t " , whereas in Curry and Pimentel 's work, i t was the most susceptible c u l t i v a r . There are several possible explanations. Morta l i ty during development might vary on the f i v e cu l t i va r s so that the number of adults emerging does not necessari ly r e f l e c t the number of 72 Table XIV. Comparison of whitef ly fecundity as measured in Experiments I and II with the whitef ly population sizes estimated 42 days a f te r indiv idual plants of f i ve tomato cu l t i va r s had been infested with 50 wh i te f l i e s by Curry and Pimentel (1972). Tomato cu l t i va r s most res i s tant most susceptible Population s ize Swi f t Roma VF Sunray Tiny Tim Campbel1 1327 (Curry & Pimentel 1972) 1194.8a* 2001 . lab 2064.1b 2314.5b 2704.9b Fecundity Campbell 1327 Swift Sunray Roma VF Tiny Tim Experiment I 73.3a** 116.0a 126.5a 272.5b 372.1c Experiment II 96.6a** 128.1a 154.7a 183.9a 195.8a *mean separation by Duncan's mult iple range te s t , 5% l e v e l . **mean separation by Student-Newman-Keuls mult ip le range te s t , 5% l e v e l . eggs deposited on a par t i cu la r c u l t i v a r . This s i tuat ion seems unl ike ly since Curry and Pimentel (1971) found no difference in such mortal i ty during development on two cu l t i va r s on which wh i te f l i e s showed very s i gn i f i can t differences in fecundity and longevity. Genetic d r i f t in d i f fe rent seed lots of the same tomato cu l t i v a r could have a substantial e f fect on an unselected character i s t i c such as glandular hair density and i t s impact on the wh i t e f l i e s . The re l a t i ve s u s cep t i b i l i t y of Tiny Tim and resistance of Swift remained stable regardless of methodology. This suggests that the s i gn i f i can t difference in mean glandular hair density between these two cu l t i va r s as measured in the present study was very l i k e l y responsible for the differences in population s izes on the two cu l t i va r s as measured 73 by Curry and Pimentel (1972). High v a r i a b i l i t y in mean glandular hair density on Sunray has been indicated in the resu l t s of Experiments I and II. In addit ion, an unusually high whitef ly ov ipos i t ion rate was recorded on one plant of Campbell 1327 in Experiment I I I. I t seems reasonable to conclude that glandular hair density in some tomato cu l t i va r s i s much more uniform than i t i s in others. This genetic d r i f t theory seems worthy of further study with regard to the discrepancies between the results of the present study and those of Curry and Pimentel (1972). The environment can influence glandular hair density. Stoner et a l . (1968) found differences in mean density between tomato plants of the same cu l t i v a r growing in the f i e l d and in the greenhouse. Also, the differences in mean density between Experiment I and II can be part ly ex-plained by differences in greenhouse and growth chamber environments during development of the plants. However, in both cases, the least and most densely pubescent cu l t i va r s were cf Tiny Tim and Campbell 1327 respect ively. The plants used by Curry and Pimentel (1972) were grown in a greenhouse under nylon-organdy cages a f ter in festat ion with wh i te f l i e s (19 days). If lowering the l i g h t intens i ty decreases glandular hair density, as Experiment II suggests, then the plants employed by Curry and Pimentel (1972) would also have had reduced glandular hair density. However, th i s s i tuat ion would not be expected to disturb the rank order of resistance of the cu l t i v a r s . 74 Curry and Pimentel (1971) report a mean longevity of 58.7 ± 3.4 days and a mean fecundity of 249.7 ± 19.1 eggs for female wh i te f l i e s on 8-week old plants of cv. Tiny Tim grown in a control led environment s imi la r to that in the present study. They also reported a mean development time of 48.4 + 0.33 days, from egg to adult with 57.4 ± 6.2% of the eggs completing development. Mean longevity of female wh i te f l i e s on Tiny Tim in both Experiments I and II was considerably lower than th i s value (44.2 and 28.2 days respect ive ly ) . Mean fecundity was higher (372.1 eggs) in Experiment I but lower (195.8 eggs) in Experiment II. Development time, as determined in Experiment V, was considerably shorter (27.9 days) with 73% of the eggs completing development. These differences could possibly be explained by differences in the age of plants (8-12 weeks vs. 4-8 weeks), d i f fe rent temperature regimes (21°C vs. 20.0-23.9°C), d i f fe rent races of whitef ly (New York vs. B r i t i s h Columbia), d i f fe rent f e r t i l i z e r regimes for the plants or d i f ferent materials employed in cage construction. Although there have been reports of resistance to whitef ly in accessions of L_. hirsutum (Gentile ejt al_., 1968; Hogenboom, ejt al_., 1974; de Ponti et a l . , 1975), the plants of th i s species examined in Experiment IVb did not exhib i t such resistance. Genti le et al_. - (1968) found that resistance in th i s species was highly var iab le, as was the density of glandular ha i r s . It seems l i k e l y that the population of L_. hirsutum examined in the present study was not s u f f i c i e n t l y large for the occurrence of the r e l a t i v e l y rare res i s tant plants. 75 The resistance of L_. peruvianum var. humifusum to wh i te f l i e s was observed during the summer of 1974 in an experimental plot of several tomato cu l t i va r s in the U.B.C. f i e l d laboratory. During that season, a l l the tomato plants on approximately 1 acre were infested with a pa r t i cu la r l y vigorous population of wh i te f ly . The Lph plants, which did not appear to have escaped i n fe s ta t i on , remained remarkably free of wh i te f ly . The fol lowing summer, a s im i l a r observation was made in the greenhouse, i . e . Lph remained free from whitef ly while several tomato cu l t i va r s in the same room supported moderately large whitefly. popula-t ions. This resistance was confirmed in Experiment IVb by the techniques employed in th i s study; only rare ly could wh i te f l i e s survive more than ten days on th i s species (as opposed to a mean l i fespan of greater than 30 days on most tomato c u l t i v a r s ) . Those wh i te f l i e s surviving longer than ten days on Lph were invar iably male. In add i t ion, the ov ipos i t ion rate decreased with the length of time spent feeding on Lph and mortal i ty during development was high (Experiment V). There was evidence that the resistance of Lph caused a skewing of the sex ra t io in favor of males in the f i r s t generation of wh i te f l i e s whose parents had been forced to l i v e on Lph before mating had occurred. Since unmated females produce haploid males parthenogenetically and mated females are capable of producing e i ther d i p l o i d females or haploid males, there are three possible explanations for the occurrence of more males: 1) mating i s i nh ib i ted , 2) the mechanism which determines whether a mated female w i l l deposit f e r t i l i z e d or un f e r t i l i z ed eggs i s 76 disturbed, or 3) there i s a d i f f e r e n t i a l mortal i ty during development. The combined results of Experiments V and Via lend support to the f i r s t hypothesis. It was when f resh ly emerged and presumably unmated wh i te f l i e s were placed'on the res i s tant plant that the most skewed sex rat ios occurred (Experiment V ia) . When eggs had been deposited by females taken from the greenhouse population without regard to age, there was no s i gn i f i cant difference in the sex ra t i o of insects which had developed on the susceptible or res i s tant plants (Experiment V). This observation suggests that there was no d i f f e r e n t i a l mortal i ty between sexes,during development and also that there was no interference in the mechanism by which mated females determine the sex of the i r progeny. These results imply that the resistance of Lph i s s u f f i c i e n t to retard i f not completely i n h i b i t the establishment of a whitef ly popula-t i on . The resistance of Lph has not yet been characterized but the fol lowing observations may be usefu l : Lph has a cha rac te r i s t i c a l l y pungent odor somewhat reminiscent of lemons. The biochemical compounds responsible for th i s odor could possibly have an a n t i b i o t i c e f fect on wh i t e f l i e s . Glandular ha i rs , though present, are sparsely d i s t r ibuted and seem pa r t i cu l a r l y f r a g i l e on th i s species. The plants are a much darker green than ordinary tomato cu l t i va r s (whitef l ies have been shown by Lloyd (1921) and MacDowal (1972) to be attracted to ye l low). The cu t i c l e of Lph plants seems thicker and less penetrable than that of tomato plants. Plate 4 shows a typ ica l plant of Lph approximately 12 weeks o l d . 77 P l a t e 4. L y c o p e r s i c o n peruvianum v a r . humifusum. Note s h i n i n e s s and dark green c o l o r o f l e a v e s , a l s o absence o f f r u i t . P e d i c e l s where f l o w e r s have a b s c i s e d s u p p l y e v i d e n c e o f s e l f - i n c o m p a t i b i l i t y . 78 Before embarking on a breeding program to develop tomato cu l t i va r s res i s tant to whitef ly employing Lph, i t would be useful to know i f races of whitef ly could be expected to develop that would thr ive on the " re s i s tant " plants. Gallun (1972) states that "races, stra ins or biotypes of insects seldom i f ever develop when tolerance or nonpreference [rather than ant ib io s i s ] i s the main mechanism of resistance inherent in the p lant . " However, Gallun et al_. (1975) imply that new biotypes or races can ar i se r e l a t i v e l y eas i l y in an insect species capable of partheno-genetic reproduction. The resistance of Lph appears to have components of both nonpreference and an t ib i o s i s . Evidence for the former comes from the f ac t that Lph was f ree of whi tef ly when there were s u f f i c i e n t su itable hosts nearby. Experiments V and VI provide evidence for an t i b i o s i s . Add i t i ona l l y , Experiment VIb indicates that wh i te f l i e s which have completed development on Lph are less f i t than those which completed development on a susceptible host. The rearing of several generations exc lus ive ly on Lph would be essential to predict any breakdown in host resistance. The i n te r spec i f i c hybr id izat ion necessary to transfer the genes con-veying resistance.from Lph" to L_. esculentum raises some d i f f i c u l t , but not insurmountable problems. The two species are un i l a t e r a l l y incompatible (L_. esculentum must be the female parent) (Hogenboom, 1972); hybr id izat ion between them has been very d i f f i c u l t to e f fect and has been successful only through the use of embryo culture (Smith, 1944). In add i t ion, because indiv idual plants of Lph are se l f - incompat ib le, a s ingle i n te r spec i f i c hybrid i s s t e r i l e (Rick and Smith, 1953; McGuire and Rick, 1954). Embryo 79 culture can again be employed to backcross the hybrid to L_. esculentum (Smith, 1944). A backcrossing program with L_. esculentum as the recurrent parent hopefully would eventually y i e l d a f e r t i l e res i s tant plant with desirable f r u i t qua l i t i e s . If the resistance factor i t s e l f i s associated with an undesirable character i s t i c e.g. the pungent odor of Lph or a substance tox ic to both wh i te f l i e s and man, or i f i t i s c lose ly l inked to genes for such a cha rac te r i s t i c , then the breeding program would be f u t i l e . In addit ion, linkage of resistance to genes cont ro l l i ng the small s ize or unpa l a t i b i l i t y of Lph f r u i t might hinder a breeding program. Nevertheless breeding programs, for disease res istance, have achieved successful res i s tant cu l t i va r s by employing parents equally unpromising with respect to agronomic performance. Lph i t s e l f has been successful ly employed in a breeding program for cur ly top virus resistance in tomatoes (Martin et al_., 1971). 80 CONCLUSION Estimates of fecundity and longevity of wh i te f l i e s have been demon-strated to vary markedly on d i f fe rent host plants within the genus Lycopersicon. There are several influences to which these estimates are sens i t ive. Whitefly fecundity and longevity have been shown to decrease with an increase in the mean glandular hair density of the tomato cu l t i v a r on which the insects are caged. A dense mat of branched trichomes, as found in Van Wert's woolly mutant, appears to v i r t u a l l y i n h i b i t whitef ly feeding and ov ipos i t ion. An unident i f ied factor found in L_. peruvianum var. humifusum d r a s t i c a l l y reduced whitef ly longevity and fecundity. The l a t t e r two types of resistance have not been previously reported with respect to wh i t e f l i e s . The pool of genetic v a r i a b i l i t y within the genus Lycopersicon seems s u f f i c i e n t l y great for the successful development of a tomato cu l t i va r res i s tant to wh i te f l i e s . 81 LITERATURE CITED Aina, O.J., J.G. Rodriguez and D.E. Knave!. 1972. Characteriz ing resistance to Tetranychus urt icae in tomato. J . Econ. Entomol. 65:641-43. Anon. 1973. Glasshouse wh i te f l y . G.B. Min i s t . Agr ic . F i sh. Food Advis. Lea f l . 86, 5 pp. Beckman, C.H., W.C. Mueller and W.E. McHardy. 1972. The l o c a l i z a t i o n of stored phenols in plant hairs . Phys io l .•Plant Pathol. 2:69-74. B lakely, R.M. and C.W. Carlberg. 1964. Swift tomato. Can. J . Plant S c i . 44:497. B r i t t o n , W.E. 1902. The wh i te - f l y or plant-house Aleyrodes. Conn. Agr ic . Exp. Stn. B u l l . (New Haven) 140:3-17. Bugbee, W.M. 1962. Whitefly transmission of Xanthomonas pe la rgon i i . Phytopathology 52:5. Burnett, T. 1949. The e f fec t of temperature on an insect host-parasite population. Ecology 30:113-34. Burnett, T. 1964. Host l a rva l morta l i ty in an experimental host-parasite population. Can. J . Zool. 42:745-65. Canerday, T.D., J.W. Todd and J.D. Dilbeck. 1969. Evaluation of tomatoes for fruitworm res istance. J . Georgia Entomb!. Soc. 4:51-54. [c i ted by Fery and Cuthbert, 1973]". Cantelo, W.W., A.L. Boswell and R.J. Argauer. 1974. Tetranychus mite repel lent in tomato. Environ. Entomol. 3:128-30.• Clayberg, CD . and J.B. Kring. 1974. Breeding tomatoes res i s tant to potato aphid and wh i te f l y . [abstract] HortScience 9:297. Clayberg., CD . 1975. Insect resistance in a graft-induced p e r i c l i n a l chimera of tomato. HortScience 10:13-15. Cockere l l , T.D.A. 1902. The c l a s s i f i c a t i o n of the Aleyrodidae. Proc. Acad. Nat. S c i . Phi l a . 54:279-83. Curry, J.P. and D. Pimentel. 1971. L i f e cycle of the greenhouse wh i te f l y , Trialeurodes vaporariorum, and population trends of the wh i te f ly and i t s paras i te, Encarsis formosa, on two tomato v a r i e t i e s . Ann. Entomol. Soc. Am. 64:1188-90. 82 Curry, J.P. and D. Pimentel. 1972. Evaluation of tomato var ie t ie s for resistance to greenhouse whi tef ly . J . Econ. Entomol. 64:333-34. de Pont i , O.M.B., G. Pet and N.G. Hogenboom. 1975. Resistance to the glasshouse whitef ly (Trialeurodes vaporariorum Westw.) in tomato (Lycopersicon esculentum M i l l . ) and related species. Euphytica 24: 645-49. Duffus, J.E. 1965. Beet pseudo-yellows virus transmitted by the green-house whi tef ly . Phytopathology 55:450-53. Engelmann, Franz. 1970. The physiology of insect reproduction. Pergamon Press, Oxford.. 307 pp. Fery, R.L. and F.P. Cuthbert, J r . 1973. Factors a f fect ing evaluation of fruitworm resistance in the tomato. J . Am. Soc. Hort ic. S c i . 98:457-59. Fery, R.L. and F.P. Cuthbert, J r . 1974a. Ef fect of plant density on fruitworm damage in the tomato. HortScience 9:140-41. Fery, R.L. and F.P. Cuthbert, J r . 1974b. Resistance of tomato cu l t i va r s to the fruitworm, Hel ioth i s zea. HortScience 9:469-70. Fery, R.L. and F.P. Cuthbert, J r . 1975. Ant ib ios i s in Lycopersicon to the tomato fruitworm (Hel iothis zea) J . Am. Soc. Hort ic . S c i . 100:276-78. Gal lun, R.L. 1972. Genetic interre lat ionsh ips between host plants and insects. J . Environ. Qual. 1:259-65. Gallun, R.L., K.J. Starks and W.D. Guthrie. 1975. Plant resistance to insects attacking cereals. Annu. Rev. Entomol. 20:337-57. Garman, H. and H.H. Jewett. 1922. The wh i te - f l i e s of hothouses. Ky. Agr ic . Exp. Stn. B u l l . 241:76-111. Gent i le , A.G. and A.K. Stoner. 1968a. Resistance in Lycopersicon and Solanum species to the potato aphid. J . Econ. Entomol. 61:1152-54. Gent i le, A.G. and A.K. Stoner. 1968b. Resistance in Lycopersicon spp. to the tobacco f l ea beetle. J . Econ. Entomol. 61:1347-49. Gent i le , A.G., R.E. Webb and A.K. Stoner. 1968. Resistance in Lycopersicon and Solanum to greenhouse wh i te f l i e s . J . Econ. Entomol. 61:1355-57. Gent i le , A.G., R.E. Webb and A.K. Stoner. 1969. Lycopersicon and Solanum spp. res i s tant to the carmine and the two-spotted spider mite. J_. Econ. Entomol. 62:834-36. Gibson, A. and W.A. Ross. 1940. Insects a f fect ing greenhouse plants. Can. Dep. Agr ic. Publ. 695, Farmer's B u l l . No. 94:37-38. 83 Gibson, R.W. 1971. Glandular hairs providing resistance to aphids in certa in wi ld potato species. Ann. Appl. B i o l . 68:113-19. G i l be r t , J . C , J.T. Chinn and J.S. Tanaka. 1966. Spider mite tolerance in mult iple disease res i s tant tomatoes. Proc. Am. Soc. Hort ic. S c i . 89:559-62. Hargreaves, E. 1915. The l i f e history and habits of the greenhouse whi te - f l y (A. vaporariorum Westw.) Ann. Appl. B i o l . 1:303-34. Harr i s , K.F. and R.H.E. Bradley. 1973. Importance of leaf hairs in the transmission of tobacco-mosaic virus by aphids. Virology 52:295-300. Haseman, L. 1946. Influence of s o i l minerals on insects. J . Econ. Entomol. 39:8-11. Hoagland, D.R. and D.I. Arnon. 1938. The water-culture method for growing plants without s o i l . C a l i f . Agr ic. Exp. Stn. C i r c . 357:1-39. Hogenboom, N.G. 1972. Breaking breeding barr iers in Lycopersicon. 1. The genus Lycopersicon, i t s breeding barr iers and the importance of breaking these bar r ie r s . Euphytica 21:221-27. Hogenboom, N.G., O.M.B. de Ponti and G. Pet. 1974. Breeding tomato (Lycopersicon esculentum M i l l . ) for resistance to whitef ly (Trialeurodes vaporariorum Westw.) Zaadbelangen 28(7):160-63. [c i ted in Plant Breeding Abstracts 45:544 (6879) 1975]. Hussey, N.W. 1958. Notes on a fungus pa ra s i t i c on greenhouse wh i te f ly . Plant Pathol. (Lond.) 7:71-72. Hussey, N.W. and B. Gurney. 1958. Greenhouse white f l y (Trialeurodes vaporariorum Westwood). Glasshouse Crops Res. Inst. Annu. Rep., 1957:134-37. Hussey, N.W., W.J. Parr and B. Gurney. 1959. The e f fec t of wh i te - f l y populations on the cropping of tomatoes. Glasshouse Crops Res. Inst. Annu. Rep., 1958:79-86. Hussey, N.W., and B. Gurney. 1960. Some host plant factors af fect ing fecundity of w h i t e - f l i e s . Glasshouse Crops Res. Inst. Annu. Rep. 1959:99-103. Johnson, B. 1956. The influence of aphids on the glandular hairs of tomato plants. Plant Pathol. (Lond.) 5:130-32. Levin, D.A. 1973. Role of trichomes in plant defense. Q. Rev. B i o l . 48:3-15. 84 Lindquist, R.K., W.L. Bauerle and R.R. Spadafora. 1972. Effect of the greenhouse whitef ly on y ie lds of greenhouse tomatoes. J . Econ. Entomol. 65:1406-08. Lloyd, LI. 1921. Notes on colour tropism of Asterochiton (Aleurodes) vaporariorum Westwood, B u l l . Entomol. Res. 12:355-59. Lloyd, LI. 1922. The control of the greenhouse wh i te - f l y (Asterochiton vaporariorum) with notes on i t s biology. Ann. Appl. B i o l . 9:1-32. Luckw i l l , L.C. 1943. The genus Lycopersicon. The University Press, Aberdeen (Scotland). 44;~pp. MacDowal, F.D. 1972. Phototactic action spectrum for whitef ly and the question of color v i s i on . Can. Entomol. 104:299-307. Martin, M.W., O.S. Cannon, and W.G. Dewey. 1971. Pedigree history of cur ly - top-res i s tant tomato l ines released by USDA. Rep, of the Tomato Genet. Cooperative 21:23-24. Mason, H.C., T . J . Henneberry and R. Lehr. 1960. Reaction of Drosophila to d i f fe rent va r ie t ie s of tomatoes. J . Econ. Entomol. 53:337-39. Maxwell, F.G., J.N. Jenkins, and W.L. Parrott. 1972. Resistance of plants to insects. Adv. Agron. 24:187.-265. McClanahan, R.J. 1970. Integrated control of the greenhouse whitef ly on cucumbers. J . Econ. Entomol. 63:599-601. McGuire, D.C. and CM. Rick. 1954. Se l f - i ncompat ib i l i t y in species of Lycopersicon sect. Eriopersicon and hybrids with L_. esculentum. Hi lgardia 23:101-24. McKinney, K.B. 1938. Physical character i s t i c s on the fo l iage of beans and tomatoes that tend to control some small insect pests. J_. Econ. Entomol. 31:630-31. McLeod, J.H. 1938. The control of the greenhouse whitef ly in Canada by the parasite Encarsia formosa Gahan. S c i . Agr ic . 18:529-35. McMullen, R.D. 1964. The t o x i c i t y of malathion, Perthane, maneb, and zineb to T. vaporariorum. Can. Entomol. 96:541-49. Mi 11 i ron , H.E. 1940. A study of some factors a f fect ing the e f f i c i ency of Encarsia formosa, an Aphelinid parasite of the greenhouse wh i te f ly . Mich. Agr ic. Exp. Stn. Tech. B u l l . No. 173, 23 pp. Minges, P.A., ed. 1972. Descriptive l i s t of vegetable va r i e t i e s . The American Seed Trade Associat ion, Inc. Washington, D.C. and The American Society for Hort icu l tura l Science, St. Joseph, Mich. 194 pp. 85 M o r r i l l , A.W. 1905. The greenhouse whitef ly (Aleyrodes vaporariorum Westw.) U.S. Dep. Agr ic. Bur. Entomol. C i r c . 57:1-9. Parr, W.J. 1968. B io logical control of the glasshouse whitef ly (J_. vaporariorum West.) by the parasite Encarsia formosa Gahan on tomatoes. Glasshouse Crops Res. Inst. Annu. Rep., 1967:137-141. Parr, W.J. 1970. Glasshouse wh i te f ly . Glasshouse Crops Res. Inst. Annu. Rep., 1969:109-110. Parr, W.J., N.E.A. Scopes, S.M. Biggerstaff, M. Ledieu and D.L. Stacey. 1975. Progress towards an integrated control programme for pests and diseases of tomatoes. Glasshouse Crops Res. Inst. Annu. Rep. 1974:89-94. Plage, R.F. 1975. Host plant resistance to the greenhouse whitef ly (Trialeurodes vaporariorum Westw.) in the tomato (Lycopersicon esculen- tum M i l l . ) and related species, espec ia l l y Solanum penne l l i i C o r r e l l . Dissertat ion Abstracts International 36:2549-50B. Quaintance, A.L. and A.C. Baker. 1914. C l a s s i f i c a t i on of the Aleyrodidae, .'• •. • Part •IT.. U:S. Dep. 'Agric. Bur.-Ehtomol: Tech. Ser. 27:95-109. Quaintance, A.L. and Baker, A.C. 1917. A contr ibution to our knowledge of the wh i te f l i e s of the family Aleyrodiriae (Aleyrodidae). Proc. U.S. Nat l . Mus. 51:335,-445. Rick, CM. 1955. Wom - a new a l l e l e of wooly. Rep, of the Tomato Genet. Cooperative 5:25-26. Rick, CM. and L. But ler. 1956. Cytogenetics of the tomato. Adv. Genet. 8:267-382. Rick, CM. and P.G. Smith. 1953. Novel var ia t ion in tomato species hybrids. Am. Nat. 87:359-73. Rodriguez, J .G. , D.E. Knavel and O.J. Aina. 1972. Studies in the r e s i s -tance of tomatoes to mites. J . Econ. Entomol. 65:50-53. Russe l l , Louise M. 1948. The North American species of wh i te f l i e s of the genus Trialeurodes. U.S. Dep. Agr ic. Misc. Publ. 635, 85 pp. Russe l l , L.M. 1963. Hosts and d i s t r i bu t i on of f i v e species of Trialeurodes (Homoptera: Aleyrodidae) Ann. Entomol. Soc. Am. 56:149-53. Schrader, F. 1920. Sex determination in the wh i te - f l y (Trialeurodes vaporariorum). J . Morphol. 34:267-305. Schrader, F. 1926;?. Notes-on the? English rand American races of the greenhouse wh i te r f l y , Ann. Appl. B i o l . 13:189-96. 86 Scopes, N.E.A. and Biggerstaff, S.M. 1971. The production., handling and d i s t r i bu t i on of the whitef ly Trialeurodes vaporariorum and i t s parasite Encarsia formosa for use in b io log ica l control programmes in glasshouses. Plant Pathol. (Lond.) 20:111-16. S h i l l i n g , P.R. 1973. Effects of the woolly locus in the tomato on amino acid and tota l protein synthesis. Bot. Gaz. 134:185-88. Smith, P.G. 1944. Embryo culture of a tomato species hybrid. Proc. Am. Soc. Hort ic. S c i . 44:413-16. Soost, R.K. 1957. Another a l l e l e of Wo. Rep, of the Tomato Genet. Coopera- t i ve 7:13. Speyer, E.R. 1927. An important parasite of the greenhouse wh i te - f l y (Trialeurodes vaporariorum, Westwood). B u l l . Entomol. Res. 17: 301-8? Speyer, E.R. 1929. The greenhouse wh i te - f l y (Trialeurodes vaporariorum Westwood). J.R. Hort ic. Soc. 54:181-92. S t o l l , N.R. and A.F. Shu l l . 1919. Sex determination in the wh i te f ly . Genetics 4:251-60. Stoner, A.K. 1970. Selecting tomatoes res i s tant to spider mites. J_. Am. Soc. Hort ic. Sc i . 95:78-80. Stoner, A.K., J.A. Frank and A.G. Gent i le. 1968a. The re lat ionsh ip of glandular hairs on tomatoes to spider mite res istance. Proc. Am. Soc. Hort ic. Sci.93:532-38. Stoner, A.K. and A.G. Gent i le . 1968. Resistance in Lycopersicon species to the carmine spider mite. U.S. Dep. Agr ic. Prod-. Res. Rep, no. 102, 9 pp. Stoner, A.K. and H.C. Mason. 1966. Attractiveness of tomato va r ie t ie s to Drosophila melanogaster. HortScience 1:89-90. Stoner, A.K. and H.C. Mason. 1969. Methods of se lect ing tomatoes for Drosophila res istance. J . Am. Soc. Hort ic . S c i . 94:396-400. Stoner, A.K. and T. Str ingfe l low. 1967. Resistance of tomato var ie t ie s to spider mites. Proc. Am. Soc. Hort ic. S c i . 90:324-29. Stoner, A.K., H.C. Mason and B.E: Otto, 1972. Resistance to Drosophila ov ipos i t ion in f i e l d plots of two tomato breeding l i n e s . J . Am. Soc. Hort ic. S c i . 97:795-96. 87 Stoner, A.K., H.C. Mason and R.E. Webb. 1969. Reaction of tomato breeding l ines to Drosophila melanogaster. HortScience 4:292-93. Stoner, A.K., R.E. Webb and A.G. Gent i le . 1968b. Reaction of tomato var iet ies and breeding l ines to aphids. HortScience. 3:77. Thomsen, M. 1925. Sex determination in Trialeurodes vaporariorum. Nature 116:428. Thurston, R., W.T. Smith and B.P. Cooper. 1966. A lka lo id secretions by trichomes of Nicotiana spp. and resistance to aphids. Entomol. Exp. Appl. 9:428-32. Trehan, K.N. 1940. Studies on the B r i t i s h wh i te - f l i e s (Homoptera, Aleyrodidae).Trans. R. Entomol. Soc. Lond. 90:575-616. Uphof, J.C.T. 1962. Plant hairs . Handb. Pflanzenanat. 4(5):l-206. van Emden, H.F. 1966. Plant resistance to insects induced by environment. Sc i . Hort ic. 18:91-102. Varma, P.M. 1963. Transmission of plant viruses by wh i t e f l i e s . B u l l . Nat. Inst. S c i . India no.24:11-33. [c i ted from Rev. Appl. Entomol. 53:105=(1965)]. Wardlow, L.R., F.A.B. Ludlam and N. French. 1972. Insectic ide resistance ' in glasshouse whi tef ly . Nature 239(5368):164-65. Webb, R.E., A.K. Stoner and A.G. Gent i le . 1971. Resistance to leaf miners in Lycopersicon accessions. J . Am. Soc. Hort ic. S c i . 96:65-67. Weber, H. 1931. Lebensweise und umweltbeziehungen von Trialeurodes vaporiorum (Westwood) (Homoptera-Aleurodina). Erster beitrag zu einer monographie dieser A r t . [Biology and environmental re lat ions of X- vaporariorum. F i r s t contr ibut ion to a monograph of th i s speciesT] Z. Morph. Oekol. Tiere 23:575-753. [ in German, English abstract in Rev. Appl. Entomol. 20:195, 1932.] Wescott, C. 1960. Plant disease handbook, 2nd ed. D. VanNostrand Co., Inc., Princeton, N.J. 825 pp. Westwood, J.O.. 1856. A. vaporariorum. Gardener's Chron. 1856:852. Wil l iams, C B . 1917. Some problems of sex rat ios and parthenogenesis. ' J . Genet. 6:255-67. Wolfenbarger, D.A. 1965. Tomato, Lycopersicon esculentum, and Lycopersicon species and genetic markers in re la t i on to mite, Tetranychus marianae, in festat ions . J . Econ. Entomol. 58:891-93. 88 Wolfenbarger, D.A. 1966. Variations in leaf miner and f l ea beetle injury in tomato va r i e t i e s . J . Econ. Entomol. 59:65-68. Yeager, A.F. and E. Meader. 1938. Short cuts in tomato breeding. Proc. Am. Soc. Hort ic. Sc i . 35:539-40. Zar, J.H. 1974. B i o s t a t i s t i c a l analys i s . Prentice H a l l , Inc., Englewood . C l i f f s , N.J. 620 pp. 89 APPENDIX I Experiment I (Table IV) ANOVA/longevity Source d.f. Mean square Error F Block 2 501.28 (C) 2.33 F2,11,0.05 3.09 Cu l t i var 4 3551.40 (A) 10.98* F4,8,0.05 3.84 Error (A) 8 323.52 Sex 1 303.20 (B) 1.32 F l ,10,0.05 = 4.96 Sex x Cul t ivar 4 105.76 (B) 0.46 F4,10,0.05 = 3.48 Error (B) 10. 229.99 Error (C) I l l 253.82 Total 140 ANOVA/oviposition rate Source d.f". Mean square F Block 2 0.08 0.10 F2,8,0.05 = 4.46 Cul t ivar 4 10.66 14.11* . F4,8,0.05 = 3.84 Error 8 .76 Total 14 ANOVA/fecundity Source d.f. Mean square F Block 2 5557.86 2.26 F2,8,0.05 = 4.46 Cul t ivar 4 47326.93 19.26* F4,8,0.06 = 3.84 Error 8 2457.59 Total 14 90 ANOVA/GHD-UE Source d.f. Mean square F Day 16 1248.20 1.39 . F l g 6 4 0 0 5 = 1 - 8 0 Cult ivar 4 14627.52 16.28t F„ '„ ' n K = 2.53 4,64,0.05 Error 64 898.49 Total 84 ANOVA/GHD-LE Source d.f. Mean square F Day 16 217.40 1.62 F ] 6 = 1.80 Cu l t i var 4 170.36 1.27 F 4 64 0.05 = 2 , 5 3 Error 64 133.95 Total 84 91 APPENDIX II Experiment II (Table V) ANOVA for longevity of wh i te f l i e s Source d.f. Mean square F Block 2 29.44 0.00 Cul t ivar 4 2297.50 (A) 9.30* Error (A) 7 247.08 0.96 Sex 1 148.17 (B) 0.65 Variety x Sex 4 242.47 (B) 1.06 Error (B) 9 229.37 .89 Error 94 258.36 Total 121 ANOVA for ovipos it ion rate Source d.f. Mean square F Block 2 1.53 2.07 F 2 8 0 05 = 4 , 4 6 Cult ivar 4 2.53 3.43 F4 Vo 05 = 3 , 8 4 Error 8 .74 Total 14 ANOVA for fecundity Source d.f. Mean square F Block 2 2952.20 1.20 F 2 8,0.05 = 4 , 4 6 Cult ivar 4 5065.50 2.05 F 4 8 0 05 = 3 , 8 4 Error 8 2468.09 Total 14 92 ANOVA for GHD-UE Source Day Cul t ivar Error Total d.f. 14 4 56 74 Mean square 774.16 6054.83 315.88 F 2.45* 19.17* 14,56,0.05 "4,56,0.05 1.86 2.53 ANOVA for GHD-LE Source d.f. Day 14 Cu l t i var 4 Error 56 Total 74 Mean square 171.14 92.69 65.70 r ' F 2.60* 1.41 F14,56,0.05 1 , 8 6 4,56,0.05 2.53 93 Experiment III (Table ANOVA/longevity Source d.f. Sex 1 Cu l t ivar 1 Fumigation 1 Sex x Cu l t ivar 1 Sex x Fumigation 1 Cu l t i var x Fumigation 1 Sex x Cu l t ivar x Fumigation 1 Error 148 Total 155 ANOVA/oviposition rate Source d.f. Cult ivar 1 Fumigation 1 Cult ivar x Fumigation 1 Error 12 Total 15 ANOVA/oviposition rate Source d.f. Plant 7 Error 8 Total 15 APPENDIX III VI) Mean square F 344.96 2. 07 4794.13 28. 81* 182.05 1. 09 180.32 1. 08 98.15 0. 59 86.18 0. 52 6.66 0. 04 166.38 Mean square F 0.025 0. 02 0.143 0. 12 0.268 0. 22 1.233 (2) (Table VII) Mean square F 1.710 4.19* .407 94 APPENDIX IV Experiment IVa (Table VIII) ANOVA/longevity Source d.f. Mean square E r r o r - - F Block 1 453.68 (C) 1.40 F l ,75,0.05 Cu l t i var 4 1686.60 (A) 4.08 F4,4,0.05 = Error (A) 4 413.47 Sex 1 406.64 (B) 1.82 F l ,5,0.05 = Sex x Cu l t i var 4 108.22 (B) 0.48 F4,5,0.05 = Error (B) 5 223.16 Error (C) 75 324.58 Total 94 ANOVA/oviposition rate Source d.f. Mean square F Block 1 0.40 0.23 F l ,4,0.05 = 7.71 Cul t ivar 4 2.41 1.38 F4,4,0.05 = 6.30 Error 4 1.70 Total 9 ANOVA/fecundity Source d.f. Mean square F Block 1 9781'. 88 5.89 F l ,4,0.05 = 7.71 Cu l t i var 4 17050.16 10.26* F4,4,0.05 = 6.30 Error 4 1662.02 Total 9 95 ANOVA/GHD-UE Source d.f. Mean square F Block 14 439.73 2.10* F ] 4 5 6 0 0 5 = 1.88 Cul t ivar 4 9700.67 46.27* F„ ' ' " = 2.54 4 }56,0.05 Error 56 209.64 Total 74 ANOVA/GHD-LE Source d.f. Mean square F Block 14 59.74 1.04 F ^ ^ ^ = 1.88 Cu l t i var 4 703.05 12.28* F, r , n n , = 2.54 4,56,0.05 Error 56 57.26 Total 74 Experiment IVb (Table IX) Since there were not the same number of blocks per c u l t i v a r , nor were the blocks simultaneous, the fol lowing data has not been analyzed for block e f f ec t . Block has rarely been a s i gn i f i cant source of variance in the i n the whitef ly data of ear ly experiments. ANOVA/longevity Source d.f. Mean square F Sex 1 514.55 4.47* F ] = 3.89 Plant material 5 3714.56 32.28* F 5 ] 8 3 0 ^ Q 5 = 2.26 Sex x cu l t i v a r 5 342.73 2.98* F 5 ' 1 8 3 ' 0 . o 5 = 2 * 2 6 Error 183 115.07 Total 194 ANOVA/oviposition rate Source d.f. Mean square F Plant material 7 39.24 22.28* F ? 1 8 0 0 5 = 2.58 Error 18 1.76 Total 25 96 ANOVA/fecundity Source d.f. Pla n t material 5 E r r o r 15 Tota l 20 ANOVA/GHD-UE (Table X) Source d .f. Pla n t m a t e r i a l 2 E r r o r 34 To t a l 36 ANOVA/GHD-LE (Table X) Source d . f . Pl a n t M a t e r i a l 2 E r r o r 34 Tota l 36 Mean square F 24035.75 18.40* F, 1 K n n r = 2.90 5,15,0.05 1306.36 Mean square 1723.50 ?1247.65 F 6.96* 2,34,0.05 = 3.28 Mean square F 120.29 50.81 312 61.41* F =3.28 97 APPENDIX V Experiment V (Table XI) ANOVA/number of f i r s t instars Source d.f. Mean square F Block 4 43.65 1.15 F 4 4 n 05 = 6 , 3 9 Lph/TT 1 57.60 1.52 F i ' 4 ' 0 . ' o 5 = 7 ' 7 1 Error 4 37.85 Total 9 ANOVA/per cent of f i r s t nymphal instars which emerged as adults. Source d.f. Mean square F B l o c k 4 T ' 0 4 5 ' 2 2 F 4 ,306 ,0 .05 = L£h/TT 1 12.39 62.27 F ^ g ^ -Error 306 0.20 Total 311 ANOVA/adults emerging " . Js Source d.f. Mean square F Block 4 50?6 3.49 F4 4 0 05 = 6 , 3 9 Lph/TT 1 562.5 38.79* F i V o ' o 5 = 7 , 7 1 Error 4 14.5 Total 9 ANOVA/proportion of males emerging Source d.f. Mean square F Block 4 0.36 1.55 F- n nc = 2.40 4,163,0.05 LpJi/TT 1 0.71 3.08 F, > 1 6 3 ^ - 3.91 Error 163 0.23 Total 168 98 Analysis of covariance/males emerging as a covariate of nymphal instars hatching and adults emerging. Source d.f. Mean square F Block 4 4.68 1.06 F 4 2 0 05 = 1 9 , 2 5 Lph/TT 1 10.67 2.42 F i 2 0 05 = 1 8 , 5 1 Error 2 4.42 ANOVA/development time Source d.f. Mean sqi Sex 1 12.32 Block 4 2.83 Lph/TT 1 86.38 Sex x Lph/TT 1 14.80 Error 161 4.82 Total 168 2 , 5 5 F l,161,0.05 = 3 , 9 1 0 , 5 9 F4,161,0.05 = 2 , 4 3 1 7 , 9 1 * F l ,161,0.05 = 3 , 9 1 3.06