INVESTIGATIONS ON THE MUTILLID WASPS 17 



which may show curves with three or even more modes. Kellogg and Bell" 

 have mentioned the possibility of bimodal or even polymodal variations. 



Dasymutilla Uoculata (Cresson) offers a good example of variation which 

 displays the bimodal characteristic. When a large number of individuals of 

 this species are arranged in a gradatory series, the curve expressing the fre- 

 quency and range of the differences in size is bimodal. The case is an interest- 

 ing one because the variation in size can be definitely attributed to an 

 ecological factor, and because it emphasizes the importance of a knowledge 

 of the ecology of an insect to a taxonomist in determining specific limits. 



DasymuiiVia bioculata is a Mutillid wasp which is exceedingly abundant in 

 the sand dune areas of Minnesota. My attention was first attracted to the 

 species in making a collection of Mutiilidae in a sand dune area 2 miles north 

 of the city limits of Minneapolis in Anoka County. The principal collecting 

 ground in this sand dune area is a blow-out (a depression blown out by the 

 wind in areas of shifting sand) about 10 acres in extent. During the season 

 of 1922, nearly 1.200 specimens of Mutillids were collected in this blow-out. 

 Other species of Hymenoptera were also abundant, particularly two species of 

 Bembicidae, Bcmhix pruinosa Fox and Microhcmlcx monodonta Say, and the 

 Sphecid wasp. Sphex argcntatus Hart. In general, the hymenopterous fauna 

 of the sand dune areas of Minnesota is very similar to that of the s:ind areas 

 of Illinois, which has been discussed by Hart (1907). The collection of 1922 

 showed that the most abundant Mutillid in this blow-out was the male Dasy- 

 mutilla bioculata (Cresson), and the female Dasymutilla cMamydata 

 (Melander), wliich was described from specimens taken in the Illinois sand 

 areas. When this material was worked over for classification, it was found 

 that the 440 specimens of the female chlamydata could be separated into two 

 series according to size, that is, those varying between 6.5 mm. and 10 mm. 

 and those varying between 11 mm. and 15 mm. The 206 specimens of the 

 male bioculata could also be separated into two similar series. No structural 

 characters could be found for separating these two series, either in the female 

 chlamydata or the male bioculata. The genitalia of the males in the two series 

 were compared, but no differences could be found. Hart (1907) first suggested 

 that chlamydata was probably the fem:ile of bioculata, but retained the name 

 chlamydata for all his specimens. During the season of 1923 specimens of the 

 female chlamydata and the male bioculata were reared from the same host and 

 were successfully mated in the laboratory. Specimens were also taken in 

 coitu in the field. Dasymutilla chlamydata Melander therefore becomes a 

 synonym of Dasymutilla bioculata (Cresson). 



Knowing that the Mutiilidae are parasitic upon wasps and bees, and tlnit 

 the species Microbembex monodohta Say and Bembix pruinosa Fox were the 

 most abu!!daut Hymenoptera in the blow-out, it occurred to me that probably 

 they were parasilized by one or more species of the Mutillids. It also seemed 

 probable that Dasyinutilla bioculata was the Mutillid concerned, parasitizing 

 either one or the other, or both of the Bembicids. 



I therefore collected 285 cocoons of Microbembex monodonta on May 29 and 

 brought them to the laboratory. Eighty-five cocoons of Bembix pruinosa were 

 collected on June 30 and brought to the laboratory. Each cocoon was placed 

 in a glass vial so that a definite record could be kept of each one. From 

 the 285 cocoons of Microbembex monodonta I obtained six males and five 

 females of Dasymutilla bioculata ranging in size from 7 to 10 mm. From the 



" Kellogg, Vernon L., and Bell, Ruby G. 1904. Studies of Variation in Insects, Proc. 

 Wash. Acad. Sci., vol. 6, pp. 203-332. 



