---

by

B. M. Pawson & R. E. Gold'

ABSTRACT

Caste differentiation in pseudergates obtained from mature colonies of Reticulitermes flavipes (Kollar), R. virginicus (Banks) and R. hageni Banks was investigated at 5 densities (10, 25, 50., 75 & 100). Caste differentiation occurred in all 3 species, and generally, soldiers formed before reproductives. More soldiers formed at higher pseudergate densities than at lower densities. Supplemental reproductives developed in all 3 species which included both secondary (nymphoid) and tertiary (ergatoid) forms. Reticulitermes virginicus produced more reproductives than did R.flavipes or R. hageni.  Additionally, more female reproductives formed than male reproductives. Reproductives formed within 3 to 4 months after separation from the founding colony, and all 3 species produced eggs that hatched. The number of eggs for the first clutch of eggs for R. virginicus, R. flavipes and R. hageni was 41, 67 and 68, respectively. It may be appropriate to rethink termite control strategies so as to not fractionate colonies during control procedures.

INTRODUCTION

The introduction and use of baits for termite control has resulted a renewed interest in basic biology. Termites continue to be major pests, and have profound impacts on structures throughout the world. Many control strategies rely on our knowledge of  termite behavior  and biology in order to produce desirable results; however, there is much to be learned about colony reproduction and development. Methods for collecting large numbers of termites have been developed (Esenther 1980, LaFage et al. 1983, Su & Scheffrahn 1986). Mass collections of termites have been used in mark-release-recapture studies to assess termite populations (Grace 1989, Jones 1990a, Su & Scheffrahn 1988). Termite colony foraging ranges have been estimated at different geographical locations (Grace 1990, Jones 1990b, Forschler 1994). Population estimates and foraging ranges have been used in termite control programs using baits to show suppression or control of subterranean termites (Su et al. 1991 & 1995).

In addition to behavior, studies of termites biology have increased. Basic termite biology studies date back to the turn of the century. Early researchers were interested in determining the termite caste system (Thompson 1918, Thompson & Synder 1920) and the interactions between termites and their intestinal protozoa (Cleveland 1924). Many studies investigated several aspects of caste differentiation and whether intrinsic or extrinsic factors caused caste differentiation (for review see DeWilde 1985). General descriptions of the different termite castes were given by Weesner (1965) and Wilson (1971). Thompson & Synder (1920) defined the caste structure for subterranean termites and included the 'third form' reproductives for Reticuliternies flavipes (Kollar), R. virginicus (Banks) and Prorhinoten-nes simplex Hagen. Although studies on the occurrence of supplemental reproductives in mature colonies have been done (Howard & Haverty 1980, Thorne & Noirot 1982, Darlington et al. 1992). factors influencing their development are not fully understood.

This study investigated caste differentiation when mature colonies are fractionated and the potential for these fractionated colonies to survive.

MATERIALS AND METHODS

Eight termite colonies located in College Station, TX were sampled monthly using bucket traps. Bucket traps are similar to the traps used by Su & Scheffrahri (1986) except a 3 inch collar from a plastic bucket is placed in the ground instead of a polyvinyl chloride (PVC) pipe. Colonies, located in woody areas, parks and around residential homes, have been sampled for 6 to 24 months. Between 2 and 7 bucket/block traps were located at each site. Once a month, one block with a high infestation of termites from each colony site was selected, removed and replaced with a new block of wood. The infested block of wood was returned to the laboratory where it was disassembled. After extracting the termites, they were sorted according to castes and counted. Direct counts were made of the number of soldiers and brachypterous nymphs. Pseudergates numbers were estimated by weighing 3 samples of 50 termites each and then obtaining a total weight of the sample.

After counts had been made, groups of 10, 25, 50, 75 and 100 pseudergates were separated from the main group. Each group was placed in 100 mm dia. x 15mm petri dishes containing pieces of moistened tongue depressors. Each density was replicated 2 times. Petri dishes were placed inside a plastic shoe box with lid containing a layer of moistened sand bovered with a piece of aluminum foil. Once a month, termites in all 5 densities were counted to determine survival and caste differentiation. Moisture was added to the wooden tongue depressors and to the sand as needed.

Species were identified using alates from each colony and the keys provided by Weesner (1965). Four of the 8 colonies studied were Reticulitermes virginicus (Banks). Three of the 8 colonies were R. flavipes (Kollar). The remaining colony was R. hageni Banks.

RESULTS AND DISCUSSION

The fractionated colonies of all 3 species showed declines in populations when held under laboratory conditions (Fig. 1). Reticulitermes hageni exhibited higher survival rates than either R. virginicus or R. flavipes, while the survival of R. virginicus and R. flavipes was similar. At the end of 7 months, the average proportion of surviving pseudergates was 62, 41, and 28% for R. hageni, R. virginicus and R. flavipes, respectively. Colonies of R. hageni and R. virginicus showed increases in population numbers at some density levels due to supplemental queens producing progeny.

Caste differentiation occurred in all 3 species (fig.2.). In general, soldiers formed prior to the development of reproductives.  The initial density of pseudergates appeared to influence the development of soldiers; generally higher densities (75 and 100) produced more soldiers than lower densities (10 and 2 5). The different species of Reticulitemies produced different proportions of soldiers. For R. hageni an average of 1, 1, 3, 3 and 4 soldiers formed at the 10, 25, 50, 75 and 100 density levels, respectively. The proportion of soldiers to develop from pseudergates was between 1 and 2% for all 5 density levels for R. virginicus and was about 1% for all 5 densities for R. flavipes.

Supplemental reproductives developed for all 3 species. Collectively, more reproductives were formed at higher densities than at lower densities (Fig. 2). At the 10 pseudergate density level, no supplemental reproductives developed; whereas, at the 75 and 100 pseudergate density levels, 29 and 28 supplemental reproductives developed, respectively. The number of reproductives that developed for R. hageni, R. virginicus and R. flavipes were different. R. virginicus colonies produced more reproductives than colonies of R. hageni or R. flavipes. Furthermore, nearly twice as many female reproductives formed as male reproductives took on two forms: secondary or nymphoid (which developrd from brachypterous nymphs) and tertiary or ergatiod (which developed from apterous nymphs) reproductives.  All females were of the tertiary form and developed from apterous nymphs. A mixture of both secondary and tertiary males were encountered.

Although many reproductives formed, only a few initiated egg productions and subsequent progeny. All 3 species were capable of forming reproductives and progeny (Figs. 3, 4 & 5). One of the oldest colonies in the study involved R. virginicus (Fig. 3). Reproductives formed during the 4th and 5th sampling periods. After another month (6th sampling period), the queen produced 18 eggs. When counts were made after 7 months, additional eggs had been laid and many had hatched as evidenced by the appearance of 1 st and 2nd instar apterous nymphs. During the 8th and final count, all eggs had hatched and 1 st, 2nd and 3rd instar apterous nymphs were present. The total number of progeny produced by the tertiary queen in not known. Mortality was apparent because a total of 41 eggs and progeny were observed during the 7th sampling period, whereas only 37 progeny (1st, 2nd and 3rd instars) were present during the 8th sampling period. In any event, the first clutch of eggs for this tertiary queen was at least 41 eggs.

With R. flavipes, a colony collected 2 months after the start of the experiment produced a reproductive pair that gave rise to progeny (Fig.4).  An identifiable queen (and king) was not present in the colony until the 4th sampling period. In addition to the king and queen, 6 eggs were observed. During the next sampling period (5th), at least 58 eggs were laid. In addition, 5 1st instar apterous nymphs were observed. After another month, a few additional eggs (at least 4) were laid. Many eggs had hatched, based on the appearance of 32 lst instar apterous nymphs. The size of the first clutch of eggs from this tertiary queen totaled at least 67 eggs.

With R. hageni, only one collection was made throughout the experimental period. The collection was made early in the study period and produced some interesting results. Caste differentiation occurred as early as the 2nd sampling period (Fig. 5). One pseudergate differentiated into a brachypterous nymph, and one pseudergate developed into a white-headed soldier. During the 3rd sampling period, the colony contained one brachypterous nymph and two soldiers. The next sampling period (4th) showed the same number of soldiers, but the brachypterous nymph developed into a secondary (nymphoid) reproductive. Changes continued during the 5th sampling period; a tertiary queen developed along with another soldier. An additional soldier, the 4th, developed durmg the 6th month. The structure of the colony during the 7th sampling period is shown in Fig. 5. Between the 6th and 7th sampling periods, eggs were laid and many had hatched. The first clutch of eggs for this tertiary queen (presumed to mated with a secondary king) was at least 68 eggs.

Field colonies of termites may be separated or fractionated from their nest colony through either physical or environmental factors. Physically, applications of termiticides around foundations of structures can split a colony of termites into smaller groups. Droughts, heavy rainfall or environmental disterbance may cause breakage of termite foraging tunnels. High winds and/or heavy rains may cause infested timber to move to new locations. The fate of these fractionated colonies depends on extrinsic  (moisture, food sources, etc.) and intrinsic (fitness, reproductive strategies, etc.) factors.

It is a well known fact that many animals are capable of adapting to laboratory cultures. Some animals readily adapt while other animals suffer bottle-necking where genetic information is lost. Termites can be easily reared in the laboratory. Bottle-necking probably does not severely affect laboratory colonies of termites because of their long generation time; however, studies show that termite survival declines as time in colony increases. Watanabe & Noda (199 1) showed R. speratus Banks had between 65 and 70% survival after 4 months. Survival of Rjlavipes and R. virginicus after 3 months averaged about 64 and 67%, respectively (Haverty & Howard 198 1). In the present study, survival ofR. virginicus, R.Jlavipes and R. hageni were similar to those reported in literature. Because reproductives began differentiating 2 to 4 months following their initial set up, it is doubtful that genetic variability was being lost.

Encountering supplemental reproductives in a colony is a fairly common occurrence (Howard & Haverty 1980, Thorne & Noirot 1982). Colonies of termites readily form reproductives once they are separated from their nest colony. Working with R. speratus, Watanabe & Noda (199 1) had colonies producing 1. 3 to 1. 5 reproductives per colony after 4 months. This is a higher rate of reproductives than was encountered in the present study. Also 'Watanabe & Noda (199 1) found that 70% of the reproductives were females, and all were tertiary forms; the males (30%) were either secondary or tertiary forms. The present study showed similar results; all females were of the tertiary form, while the males were either secondary or tertiary forms. Howard & Haverty (1980) also encountered more female reproductives than male reproductives when sampling a mature field colony of R. flavipes

The reproductive potential of fractionated colonies is in need of further exploration. It appears that population growth of fractionated surpasses that of founding colonies. In studies of founding pairs, Beard (1974) found that  the inital egg clutch size to be between 10 and 14 eggs. Founding colonies of R. flavipes in our laboratory have produced between 1 and 18 eggs for their first clutch (unpublished data). At the end of one year, an average of 54 progeny were produced by founding pairs (Beard 1974). In the present study, tertiary queens of R. virginicus, R. flavipes and R. hageni produced at least 41, 67 and 68 progeny during the initial egg deposition period, respectively. As a result, it appears the reproductive potential for supplemental queens (at least tertiary forms) is greater than the reproductive potential of founding queens. Birth rates of supplemental reproductives and founding pairs need to be studied over longer periods of time to determine fecundity values. It may be appropriate to rethink termite control strategies so as to fractionate colonies.

The role of male reproductives in a colony is suspect. It has been hypothesized in the past by Miller (1969) and Nutting (1969) that parthenogenesis may occur within colonies of termites. Parthenogenesis appears to have occurred during the present study. Some colonies produced tertiary females that laid eggs and hatched. Males were not observed in these colonies. Whether males were not identified, or whether they developed, mated, died and were consumed by their nest mates is unknown. Studies are presently being conducted to detennine if subterranean termites can reproduce parthenogenetically.

REFERENCES

Cleveland, L.R. 1924. The physiological and symbiotic relationships between the intestinal protozoa of termites and their host, with special reference to Reticulitermes flavipes  (Kollar). Biol. Bull. 46:203-223.

Darlington, J.P.E.C., E.M. Cancello, & O.F.F. Souza. 1992. Ergatoid reproductives in termites of the genus Dolichorhinotermes (Isoptera, Rhinotermitidae). Sociobiology. 20:41-47.

DeWilde, J. 1985. In Memoriam Martin Luscher  (1917-1979): His contributions to insect physiology and sociobiology. Pp. 3-13 In Caste Differentiation in Social Insects. Current Themes in Tropical Science, Vol. 3 (J.A.L. Watson, B.M. Okot-kotber & Ch. Noirot, eds.). Pergamon Press.

Forschler, B.T. 1994. Fluorescent spray paint as a topical marker on subterranean termites (Isoptera: Rhinotermitidae). Sociobiology. 24:1-12.

Grace, J.K. 1989. A modified trap technique for monitoring Reticulitermes Subterranean termite populations (Isoptera: Rhinotermitidae). Pan-Pac. Entomol. 65:381-384.

Grace, J. K. 1990. Mark-recdpture studies withReticulitermes flavipes (Isoptera: Rhinotermitidae). Sociobiology. 16:297-303.

Haverty, M-1. & R.W. Howard. 1981. Production of soldiers and maintenance of soldier proportions by laboratory experimental groups of Reticulitermes flavipes (Kollar) and Reticulitermes virginicus (Banks) (Isoptera: Rhinotermitidae). Ins. Soc. 28:32-39.

Howard, R.W. & M.I. Haverty. 1980. Reproductives in mature colonies of Reftculitemiesflavipes: Abundance, sex-ratio, and association with soldiers. Environ. Entomol. 9: 458-460.

Jones, S.C. 1990. Delineation of Heterotemies aureus (Isoptera: Rhinotennitidae) foraging territories in a Sonoran desert grassland. Environ Entomol. 19: 1074-1054.

LaFage, J.P., N.-Y. Su, M.J. Jones & G.R. Esenther. 1983. A rapid method for collecting large numbers of subterranean termites from wood. Sociobiology. 7: 305-309.

Miller, E.M. 1969. Caste differentiation in the lower termites. Pp. 283-310 In The Biology of Termites, Vol. I (K. Krishna & F.M. Weesner, eds.). Academic Press, New York.

Nutting, W.L. 1969. Flight and colony foundation. Pp. 233-282 In The Biology of Termites, Vol. 1 (K. Krishna & F.M. Weesner, eds.). Academic Press, New York.

Su, N.-Y. & R.H. Scheffrahn. 1986. A method to access, trap, and monitor field populations of the formosan subterranean termite (Isoptera: Rhinotermitidae) in the urban environment. 12:299-304.

Su, N.-Y. & R.H. Scheffrahn. 1988. Foraging population and territory of the formosan subterranean termite (Isoptera: Rhinotermitidae) in an urban environment. Sociobiology 14:353-359.

Su, N.-Y., P.M. Ban & R.H. Scheffrahn. 1991. Suppression of foraging populations of the formosan subterranean termite (Isoptera: Rhinotermitidae) by field applications of a slow-acting toxicant bait. J. Econ. Entomol. 84:1525-153 1.

Thompson, C.B. 1918. Origin of the castes of the common termite, Leucotemies flavipes (Kollar). J. Morphol. 30:83-153.

Thompson, C.B. &T.E. Snyder. 1920. The'third form,'the wingless reproductive type of termites: Reticulitemies and Prorhinotennes. J. Morphol. 34:59 1 - 633.

Thorne, B.L. & C. Noirot. 1982. Ergatoid reproductives in Nasutitermes corniger (Motschulsky) (Isoptera: Termitidae). Int. J. Insect Morphol. & Embryol. 11:213-226.

Watanabe, H. & H. Noda. 199 1. Small-scale rearing of a subterranean termite, Reticulitermes speratus (Isoptera: Rhinotermitidae). Appl. Ent. Zool. 26:418420.

Weesner, F.M. 1965. The termites of the United States - A Handbook. Nat. Pest Control Assoc., Elizabeth, New Jersey. 70 pp.

Wilson, E.O. 1971. Caste Termites. Pp. 183-196 In The Insect Societies. Belknap Press, Cambridge, Mass