314 Florida Entomologist 84(2) June 2001

A MORPHOLOGICALMEANS OF DISTINGUISHING FEMALES OF THE CRYPTICFIELD CRICKETSPECIES, GRYLLUSRUBENS AND G. TEXENSIS (:)

DAVID A. GRAY 4, THOMAS J. WALKER2, BRENDA E. CONLEY' AND WILLIAM H. CADE' 'Department of Biological Sciences, Brock University, St. Catharines, Ontario, L2S 3A1, Canada

2Department of Entomology and Nematology, University of Florida, Gainesville, FL, 32611-0620, USA

3Present Address: The President's Office, The University of Lethbridge, Lethbridge, Alberta, TIK 3M4, Canada

4Present Address: Department of Biology, The University of Lethbridge, Lethbridge, Alberta, TIK 3M4, Canada

The field crickets rubens Scudder and lection locality alone (G. rubens: Gainesville, FL G. texensis Cade and Otte (previously G. integer (n =15); G. texensis: Austin, TX (n = 14), Dallas, from Texas, see Cade & Otte (2000)) are the most TX (n = 6)) or based on the songs of their brothers commonly collected crickets throughout the (G. texensis: Uvalde, TX (n = 2), Austin, TX (n = southeastern United States from North Carolina 1)). Females from sympatric sites (G. rubens: Mil- to Texas. The two are sympatric from ton, FL (n = 8), Marianna, FL (n = 12), Pensacola, western Florida to eastern Texas (Walker 1998). FL (n = 5), Mobile, AL (n = 1), Decatur, AL (n = 1); These two species are the only known trilling field G. texensis: Milton, FL (n = 32), Starkville, MS (n crickets in the southeastern US, and are cur- = 9), Tuscaloosa, AL (n = 4), Pensacola, FL (n = 4), rently separated by song differences alone. Male Mobile, AL (n = 2), Carrollton, GA (n = 6)) were G. rubens produce trilled calling song with pulse identified to species based on the songs of their rates averaging ca. 55 pulses per second (p/s) at brothers. 250C, whereas G. texensis produce trills with The crickets were measured as two replicates; pulse rates averaging about 80 p/s at 25?C (Sour- replicates 1 and 2 represent crickets in the collec- oukis et al. 1992, Walker 1998, 2000 Martinet al. tions of DAG and the Florida State Collection of 2000). The pulse rates of the two species become , Gainesville, respectively. The data more similar at lower temperatures, but provided are presented together in Figure 1 and both sepa- that the temperature is above approximately rately and pooled in Table 1. We found that female 200C, males can unambiguously be identified to G. rubens have ovipositors ca. 2.5 to 3 mm longer species by song (Walker 1998; Gray & Cade 2000). relative to their body size than do female G. texen- No morphological means of distinguishing either sis. For the pooled data, we tested the effect of males or females has previously been reported species with pronotal width as a covariate using (see Nickle & Walker (1975) for other species of an ANCOVA. The model r was 0.88 (i.e., 88% of the southern United States). Here we report that the variation in ovipositor length was accounted = most females of the two species can be separated for). Both species (F,1119) 170.96, P < 0.0001) and on the basis of body-size relative ovipositor pronotal width (F1 9, = 440.93, P < 0.0001) were length: G. rubens have longer ovipositors relative significant predictors. A comparison of the slopes to body size than do G. texensis. of ovipositor length on pronotal width was made We compared the ovipositor lengths and the by testing for a species x pronotal width interac- pronotal widths of 122 females from several local- tion in a separate ANCOVA. The slopes did not ities across the species' geographic ranges. Fe- differ (G. rubens, slope = 2.285, G. texensis, slope males were either laboratory reared from field- = 2.374; F,19, = 0.10, P < 0.7507). The intercepts caught nymphs, laboratory reared offspring of of the lines were 1.34 mm for G. rubens and -1.31 field-caught field-inseminated females, or were mm for G. texensis. Thus it should be possible to field caught females. Females from allopatric separate accurately most females of these two sites were identified to species on the basis of col- species.

TABLE 1. MEAN ?SD OVIPOSITORLENGTHS (MM) OF G. RUBENS AND G. TEXENSIS.

Replicate G rubens G. texensis t (df P -value

1 14.42 ? 0.79 (n = 24) 11.46 ? 0.77 (n = 31) t(53) = 13.91 <0.0001 2 14.7 ? 2.0 (n = 18) 11.2 ? 2.4 (n = 49) t "65 = 5.53 <0.0001 Combined data 14.53 ? 1.44 (n = 42) 11.30 ? 1.91 (n = 80) t(12W =9.60 <0.0001

This content downloaded from 130.166.34.126 on Wed, 09 Sep 2015 23:36:15 UTC All use subject to JSTOR Terms and Conditions Scientific Notes 315

17.5 A~~~~~~ 16.5

15.5 A 0 A A A A~~~~ A A Ah 0 14.5 A t *A0 A A 13.5 A A~~~~~~~~~~,AA,A 0 0

A A A A 12.5 A A A

A A ? t A AA o 92.5 AA A A~~~~~~ A AIA A5t ^ i A 81.5 AA~~~~~~A A

7.5 I i I 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 Pronotal width (mm)

Fig. 1. Ovipositor length as a function of body size (pronotal width) in (filled symbols) and G. tex- ensis (open symbols). Triangles represent individuals identified to species based on the songs of their brothers, whereas circles represent individuals identified to species based on locality alone. G. rubens have longer ovipositors than G. texensis.

SUMMARY MARTIN,S. D., D. A. GRAY,AND W. H. CADE.2000. Fine- scale temperature effects on calling song. Females of the field crickets Gryllus rubens Can. J. Zool. 78: 706-712. and G. texensis can generally be distinguished on NICKLE,D. A., ANDT. J. WALKER1975. A morphological the basis of their ovipositor length relative to key to field crickets of Southeastern United States body size. G. rubens females have longer oviposi- (Orthoptera: Gryllidae: Gryllus). Florida Entomol. 57: 8-12. tors than do G. texensis. SOUROUKIS,K., W. H. CADE,AND G. ROWELL.1992. Fac- tors that possibly influence variation in the calling REFERENCES CITED song of field crickets: temperature, time, and male size, age, and wing morphology. Can. J. Zool. 70: 950- CADE,W. H., ANDD. OTTE.2000. n. sp.: 955. A widely studied field cricket (Orthoptera; Gryllidae) WALKER, T. J. 1998. Trilling field crickets in a zone of from the southern United States. Trans. Am. Ento- overlap (Orthoptera: Gryllidae: Gryllus). Ann. Ento- mol. Soc. 126: 117-123. mol. Soc. Am. 91: 175-184. GRAY,D. A., AND W. H. CADE. 2000. Sexual selection WALKER,T. J. 2000. Pulse rates in the songs of trilling and speciation in field crickets. Proc. Natl. Acad. Sci. field crickets (Orthoptera: Gryllidae: Gryllus). Ann. USA. 97: 14449-14454. Entomol. Soc. Am. 93: 565-572.

This content downloaded from 130.166.34.126 on Wed, 09 Sep 2015 23:36:15 UTC All use subject to JSTOR Terms and Conditions