(w/w) males. Thus these host females proved to have been thelygenic ( F ' + If+) with implanted pole cells from arrhenogenic (fw/fw) donors. Eggs derived from these implanted pole cells, in spite of the constitution of their hosts as carriers of the F ' sex realizer, therefore developed to males, following the genetic constitution of the donor. This indicates that the F' gene product in C. rufiJacies is synthesized in germ line cells itself. Recently in Drosophila germ line autonomy of four maternal effect mutants has been shown by pole cell transplantations [6]. The F ' gene of C. rufzfacies is the first sex realizer with maternal effect analyzed by pole cell transplantations and behaves as germ-line-autonomous. Supported by the Deutsche Forschungsgemeinschaft; the author thanks Prof. K. Ill-
mensee (Genf) for the introduction into the pole cell transplantation technique, R. Niedermtiller and M. Martensen for careful technical assistance. Received March 10, 1982 1. Ullerich, F.-H.: Mol. gen. Genet. 125, 157 (1973) 2. Ullerich, F.-H.: Chromosoma (Berl.) 50, 393 (1975) 3. Ullerich, F.-H. : Wilhelm Roux' Arch. 188, 37 (1980) 4. Ullerich, F.-H.: Entom. Gem 7, 113 (1981) 5. Illmensee, K.: Wilhelm Roux' Arch. 171, 331 (1973) 6. Marsh, J.L., et al.: Devel. Biol. 56, 195 (1977); Marsh, J.L., Wieschaus, E.: ibid. 60, 396 (1977); Wieschaus, E., Marsh, J.L., Gehring, W.J.: Wilhelm Roux' Arch. 184, 75 (1978); Regenass, U., Bernhard, H.P.: Mol. gem Genet. 164, 85 (1978)
Table 1. Recruitment response (attraction and following for at least 5 cm) to fresh, 5 min, and 10 min old artificial trails prepared from worker (LW3) or soldier sternal glands. Counts were taken in a 3-rain period. 10 replicates/each trail age. N total number of termites recruited in 10 replicates, SD standard deviation. The response to a soldier sternal gland trail was by 98.3% soldiers and 1.7% workers (fresh trail) and 100% soldiers (5- and 10-min trails). The response to a worker sternal gland trail was by 72.4% soldiers and 27.6% workers (fresh trail), 89.8% soldiers and 10.2% workers (5-min trail) and 96% soldiers and 4% workers (10-rain trail) Trail age
Gland origin Soldier
Worker
N
N
Mean-+SD
Fresh 118 11.8_+10.6 5min 40 4.0_+ 3.7 10min 12 1.2+ 1.7
Recruitment and Orientation Components in a Termite Trail Pheromone J.F.A. Traniello Department of Biology, Boston University, Boston, Massachusetts 02215, and Harvard University, Museum of Comparative Zoology Laboratories, Cambridge, Massachusetts 02138 In all species of termites studied thus far, the sternal gland is the sole source of trail pheromone. Sternal gland secretions mediate communication during recruitment to food [1, 2], and are involved in colony defense [1, 3]. Sternal gland pheromone has a mass communication function, therefore, it chemically codes for both excitation and orientation. Several studies atempting to isolate termite trail pheromones identify a single component, or several structurally related isomers, such as neocembrene A [4], hexanoic acid [5], and dodecatrienol [6]. Our recent behavioral studies on communication during foraging in Nasutiterrnes costalis indicate that the trail pheromone of this species is characterized by an ephemeral component that regulates recruitment and an incredibly persistent substance that provides a longlasting orientation cue. To demonstrate the different effects of the two components, sternal glands were dissected from workers and soldiers of N. costalis and crushed in 100% ethanol. Gland extracts (one gland/10 gl ethanol) were then used to create artificial trails 25 cm Naturwissenschaften 69 (1982)
in length, which were placed in close proximity to groups of termites. As has been amply documented in the literature, termites became excited when exposed to the trail odor and many oriented along the entire length of the trail. Detailed observations indicate that individuals do ~not appear to reinforce the artificial trail while following it. Despite possible complications due to termites depositing additional pheromone, the recruitment effect, strong at first, decayed rapidly within 10rain (Table 1). Trails prepared from third instar large workers (LW3) [7] had a stronger recruitment effect than soldier trails. Also, the response to soldier trails was highly variable, perhaps reflecting individual differences. Soldier sternal glands range in volume from 2.04-6.43 ~tm3. Although aged trails elicited only a weak recruitment response and in the case of soldier trails were nearly completely incapable of recruiting workers, individuals of both castes would still use the trails as orientation guides. To demonstrate this, trails prepared from glands dissected from soldiers and workers were aged from
9 Springer-Verlag 1982
Mean_+SD
181 18.0_+4.2 i18 11.8+_4.1 50 5.0-+3.7
10 min to 4 h and were placed at the terminus of a 25 cm long freshly prepared worker trail. As a control, a solvent trail (ethanol) was placed at a 90~ angle to the aged trail being tested. Control trails were never followed. Table 2a shows that trails aged up to 4 h still provide orientation cues although they have no recruitment effect. The orientation effects of worker trails were not different from those of soldier trails. Foragers of N. costalis travel from their arboreal nest to food sources in covered shelter tubes, or galleries, semicircular in cross section [8]. The floor of these covered trails, approximately 14 mm wide, is composed of fecal material impregnated with trail pheromone, commonly called pav~ [9]. Because these trails remain in use for extended periods of time during resource exploitation, it is possible that once a group of foragers has been recruited, the short-lived excitatory component of the trail pheromone decays, and the mobilization of foragers from the nest is turned off. Foragers are then guided by the persistent component of the trail substance. However, pay6 could have additional constituents that might serve as orientation cues, such as fecal material, which has such a function in ants [10]. We prepared artificial trails from termite rectal fluid and assayed for their effectiveness in orientation in the manner described previously. The 343