June, 2002 Journal of Vector Ecology 39 The effects of diet upon pupal development and cocoon formation by the cat flea (Siphonaptera: Pulicidae) W. Lawrence and L. D. Foil Department of Entomology, Louisiana State University Agricultural Center, 402 Life Sciences Building, Baton Rouge LA 70803-70 USA Corresponding Author Received 29 April 200; Accepted 0 July 200 ABSTRACT: Cocoon formation by cat flea larvae was directly related to the quantity of eggs or yeast consumed. Larvae consuming either -3 eggs or 0.25-.0 mg of yeast developed as naked pupae or formed incomplete cocoons. Third instar cat flea larvae fed upon naked pupae and pupae within partial cocoons, but complete cocoons protected pupae from cannibalism. First and second instars did not attack pupae. When larvae were provided with a carpet fiber for protection, a greater number of fleas successfully developed to the adult stage. Journal of Vector Ecology 27(): 39-43. 2002. Keyword Index: Cat fleas, pupae, cocoon, cannibalism. INTRODUCTION Prior to pupation, third instar cat flea, Ctenocephalides felis (Bouché), larvae spin silken cocoons. Subsequent development from pupa to adult takes approximately seven days, but adult cat fleas may remain sequestered within the cocoon until they detect stimuli such as pressure and heat (Silverman and Rust 985). The primary function of the flea cocoon appears to be protection from predation rather than against desiccation (Mellanby 933, Silverman et al. 98). Silverman and Appel (984) conducted a study on the predation of cat flea pupae by Argentine ants, Iridomyrmex humilis (Mayr), and showed that pupae within cocoons were protected, while naked pupae and pupae in damaged cocoons were readily preyed upon. Cat flea larvae are cannibalistic and readily feed upon eggs (Lawrence and Foil 2000). Thus, the cat flea cocoon could provide protection for pupa from predation by flea larvae. The purpose of this investigation was to determine the role of larval diet relative to cocoon formation and the function of the cocoon structure with respect to protection from cannibalism by flea larvae. MATERIALS AND METHODS Flea eggs (W 24 h) were obtained from a colony maintained at the Louisiana State University Agricultural Center St. Gabriel Research Station, St. Gabriel, LA (Henderson and Foil 993). Eggs that were used as a food source were held at -5 C for 24 hours and were designated as frozen eggs (FE). The larvae were collected as previously described (Lawrence and Foil 2000). Individual spirals of feces (CHF) from adult fleas maintained on cats were aspirated from the debris beneath the cat cages. The artificial larval diet contained ground dog food, whole beef blood, and brewer s yeast (Henderson and Foil 993). All assays were conducted in an incubator (Precision Model 85) at 24.0 ± 2.0 C and in constant darkness at 75% relative humidity (Sweetman 933). Egg and yeast consumption versus cocoon formation Individual first instar larvae were held in 6 x 00 mm sterile test tubes and provided with either 0,, 3, 6, 2, or 24 FE plus CHF ad libitum, 30 replicates each. The life stage and/or presence of a silk cocoon was recorded every other day from day 8 until day 24. After day 24, the larvae were examined every fifth day up to day 50. Stock suspensions of 200, 50, 00, 50, 25, and 0 mg of brewer s yeast/ ml of distilled water were prepared and 0 µl of each suspension was added to the bottom of 30 individual test tubes. After the tubes had dried, 0 spirals of CHF and a first instar larva were added to each tube. Mortality, pupation, and cocoon formation of the larvae were recorded every third day for 24 days.
40 Journal of Vector Ecology June, 2002 Cocoons were assigned a relative rating of to 3. When the developing pupae were completely visible but silk strands were present, the number was assigned. Pupae that were partially visible were rated as 2 and pupae completely covered by cocoons were rated as 3. Yeast or egg consumption versus development First, second, and third larval instars, previously reared on CHF only, were placed in 6x00 mm tubes and provided with CHF plus either yeast or FE for 24 hours (six treatments plus a control, 30 replicates each). After 24 hours, the larvae were then transferred to individual tubes containing CHF and examined every fifth day for 40 days. The control group of st instars received CHF only. Cannibalism of naked pupae and pupae within cocoons Naked pupae (NP) were obtained by placing newly hatched larvae individually into sterile 6x00 mm tubes containing mg of brewer s yeast and 0 spirals of CHF. Silk and silk plus sand cocoons were obtained by placing first instars into sterile 6x00 mm tubes containing 2 mg of yeast and 0 spirals of CHF with or without sand. After day 2, excess yeast was removed and first, second, or third instars were placed individually into the tubes that contained CHF and either a naked pupa, a silk cocoon, or a sand cocoon (30 replicates of each). One additional group of third instars was provided with a NP, five FE, and CHF. Observations were conducted each day until the first or second instar molted or until the third instars pupated. The influence of a carpet fiber upon survival The outside edges of eight 60 X 5mm glass petri dishes were moistened and larval diet was added. Then excess diet was removed after drying. A.43cm fiber of white carpet was secured to the center of four dishes. Fifty larvae were added to each of the eight dishes and were held for 24 d. Statistical analysis The effects of the various diets were compared using the Chi-square test. Regression analysis was used to determine the relationship between egg consumption and cocoon formation. Correlation analysis was used to relate egg and yeast consumption to larval development and cocoon formation. An ANOVA was used to analyze the carpet fiber assay (SAS 987); p > 0.05 was the accepted level of statistical difference. RESULTS There were no significant differences in pupal development for larvae provided with at least one FE (Table ). The larvae provided with 24 FE consumed an average of 23.27 ± 0.954 eggs. Cocoon formation was significantly different among treatments Chisquare critical value.07 < 24.87), and cocoon development increased as the number of eggs consumed increased (r =0.857). Larvae provided with 6, 2, and 24 FE pupated within 8- days, but third instars remained in the other treatments at day 24 and no further development was observed between day 24 and day 50 (Table ). At day 24 there were more larvae in the treatment groups provided with 0, 0.0, and 0.25 mg yeast when compared to the other groups (Chi-square critical value 3.84< 0.024). There were significant differences in the number of pupae and in cocoon formation among treatment groups (Chi-square critical value 4.0 < 49.09; Table 2). There was a positive correlation (r=0.903) between cocoon formation and the amount of yeast provided. Yeast or egg consumption versus development There were no differences in adult development between third instar groups provided with either FE or yeast (Chi-square critical value 3.84 > 2.63) or from larvae provided with yeast for 24 hours as st, 2nd, or 3rd instars (Chi-square critical value 5.99 > 0.90). Adult development was not different between larvae provided only CHF and those provided eggs for 24 h as first or second instars (Table 3). Table. The effects of flea egg consumption upon development and cocoon formation of first instar flea larvae maintained individually for 24 days at 24 C and 75% relative humidity and provided with frozen flea eggs plus flea feces ad libitum. No. Eggs No. Pupae No. Cocoons % Eclosion 0 8a 4a 00.0 22b 2 9a 86.4 3 29b 3 2b 86.2 6 30b 4b 96.7 2 30b 26b 93.3 24 30b 24b 96.7 4 third instars remaining after 24 days. 2 7 third instars remaining after 24 days. 3 third instar remaining after 24 days. Values within columns followed by the same letter are not significantly different; n= 80; n i = 30.
June, 2002 Journal of Vector Ecology 4 Table 2. The effects of brewer s yeast upon the development of first instar flea larvae provided with flea feces ad libitum and maintained at 24 C and 75% relative humidity for 24 days. Pupae Yeast (mg) Larvae No. Pupae Cocoon Silk No. Dead 0 20a 0a 0a NA 0 0.0 2a 3a 0a NA 6 0.25 a 7b 2a.0± 0.0 2 0.50 5b 20b 3a.3 ± 0.6 5.00 2b 23bc 8a 2.4 ± 0.9 5.50 0b 28cd 8a.5 ± 0.9 2 2.00 0b 30d 6ab.6 ± 0.9 0 4.00 2b 28cd 28b 3.0 ± 0.0 0 Relative rating on a scale of to 3. Values within columns followed by the same letter are not significantly Table 3. The development of first, second, and third instars (cat flea) provided with a diet of flea feces (CHF) plus either eggs or yeast for 24 hours and then provided CHF ad libitum for 40 days at 24ºC and 75% relative humidity. Day Instar CHF+ Larvae Pupae Adults CHF 0 0 0a Eggs 3 0 2a Yeast 5 3 8c 40 2 Eggs 6 3 0a Yeast 5 3 20c 3 Eggs 3 0 4bc Yeast 0 0 24cd Control st instars were provided with CHF only for 40 days. Values within columns followed by the same letter are not significantly different; n=20, n i =30. Table 4. The development of third instar flea larvae held individually for 24 days at 24 C and 75% relative humidity and provided with flea feces ad libitum plus either a naked pupae, a naked pupae plus 5 frozen flea eggs (FE), or a sand or silk cocoon. Developmental Stage CHF+ Larvae # of Pupae Consumed Adult A Adult B 2 Silk Cocoon 26 0 29 0 Sand Cocoon 29 0 30 0 NP 0 30 0 30 NP+FE 0 30 0 28 Developed from pupae; 2 Developed from larvae n= 240, n i = 30.
42 Journal of Vector Ecology June, 2002 Cannibalism of pupae First and second instars did not feed upon naked pupae and did not enter the silk and sand cocoons. Thirty adults emerged from the pupae in each treatment. Third instars did not penetrate either silk or sand cocoons. All naked pupae were consumed by 3rd instars within 24 hours, and 00% of the 3rd instars with naked pupae subsequently developed into adults (Table 4). Third instars provided with FE and naked pupae consumed both, and 93.3% of the larvae developed into adults. The influence of a carpet fiber The mean number of adults that developed in petri dishes without a carpet fiber (3.25 ± 9.49) was significantly different from the number that developed with a carpet fiber ( 32.00 ± 6.98). The mean number of females and males in units without a carpet fiber was 6.75 ± 4.72 and 6.50 ± 4.93, respectively; the mean number of females and males in units with carpet was 2.25 ± 2.75 and 9.75 ± 4.65, respectively. DISCUSSION The percentage of the fleas with cocoons increased as the number of consumed eggs increased. Similarly, larvae provided with more yeast constructed more complete cocoons. These data indicate that there is a direct relationship between cocoon formation and egg or yeast consumption and support the suggestion that undernourished cat flea larvae may produce weaker cocoon structures (Silverman and Rust 985). These findings also are consistent with the observations of Chippendale (978) who indicated that the nutrition obtained during the larval stage of insects (including some members of Siphonaptera) was critical to the development of silken cocoons. Consumption of eggs or yeast was required for maximal pupal development of larval cat fleas. Under the conditions of this study, only third instars consumed eggs while all three instars consumed yeast. Third instars did not eat pupae in well constructed silk and sand cocoons but 00% of the third instar larvae fed upon naked pupae with or without FE available. First and second instars did not cannibalize naked pupae or pupae within complete silk cocoons. These data indicate that the third instar is the primary cannibal and that cocoon structures can protect pupae from cannibalism by third instars. The provision of a single carpet fiber increased the percentage of larvae successfully developing into adults. These data indicate that escape from predation within substrates prior to cocoon formation may be an important survival strategy for fleas. The data from this and our previous study (Lawrence and Foil 2000) indicate that there is intense competition within flea populations. These data show that cannibalism of pupae and eggs is a normal component of cat flea behavior, and that cannibalism may serve as a population control mechanism. The single carpet fiber assay demonstrated the significance of the need for larval protection in order to successfully complete their development, as the prepupae are susceptible to predation. Dryden and Smith (994) reported that mechanical stimuli were needed for successful cocoon formation. The results of this study indicate that cocoon formation also depends upon the type and quantity of larval dietary components. Larvae that consume eggs or yeast produce cocoons that can provide protection from cannibalism by flea larvae; the third instar is the primary predator. During the period before the cocoon is completed, cat flea pupae are susceptible to predation. Our results show that substrates, such as carpet, can afford protection against cannibalism. Acknowledgments We thank Kyle Soileau for technical assistance. Published with approval of the Director of the Louisiana Agricultural Experiment Station as Manuscript No. 00-7-050. REFERENCES CITED Chippendale, G.M. 978. The function of carbohydrates in insect life processes. pp -55. In: M. Rockstein, ed., Biochemistry of Insects. Academic Press, NY. Dryden, M. and V. Smith. 994. Cat flea (Siphonaptera: Pulicidae) cocoon formation and development of naked pupae. J. Med. Entomol. 3: 272-277. Henderson, G. and L. D. Foil. 993. Efficacy of diflubenzuron in simulated household and yard conditions against the cat flea Ctenocephalides felis (Bouché) (Siphonaptera: Pulicidae). J. Med. Entomol. 30: 69-62. Lawrence, W. and L. Foil. 2000. The effects of flea egg consumption on larval cat flea (Siphonaptera: Pulicidae) development. J. Vector Ecol. 25: 98-0. Mellanby, K. 933. The influence of temperature and humidity on the pupation of Xenopsylla cheopis. Bull. Entomol. Res. 24: 97-203. SAS Institute. 987. SAS system for elementary statistical analysis. SAS Institute, Cary, N.C. Silverman, J. and A. G. Appel.984. The pupal cocoon of the cat flea, Ctenocephalides felis (Bouché)(Siphonaptera: Pulicidae): barrier to ant
June, 2002 Journal of Vector Ecology 43 predation. Proc. Entomol. Soc. Wash. 86: 660-663. Silverman, J., M. K. Rust, and D.A. Reierson. 98. Influence of temperature and humidity on survival and development of the cat flea, Ctenocephalides felis (Siphonaptera: Pulicidae). J. Med. Entomol. : 78-83. Silverman, J. and M. K. Rust. 985. Extended longevity of the pre-emerged adult cat flea (Siphonatera: Pulicidae) and factors stimulating emergence from the pupal cocoon. Ann. Entomol. Soc. Am. 78: 763-768. Sweetman, H. 933 Studies of chemical control of relative humidity in closed spaces. Ecology: 4: 40-45.