MORPHOLOGICAL EFFECTS OF CROWDING 325 



individuals of both types show that there is no interval between 

 figures for L. migratoria and for L. danica. 



In L. migratoria both sexes are about the same size; the males 

 are about 4 per cent smaller than the females. On the other hand, 

 L. danica males are about 20 per cent smaller than females of that 

 species. In both, the proportions between different parts of the 

 body remain constant, independent of the absolute measurements. 



The coloration of adults of the two forms is variable both in 

 color and in pattern, so that they cannot be separated accurately 

 on color characters. In general, L. migratoria is less variable in 

 color than is L. danica; the color markings tend to be less sharp, 

 and the general coloration tends to be more uniform. The hind 

 tibiae are usually yellowish, though occasionally they are red. 

 Locusta danica is more variable in adult coloration: bright-green 

 forms are common; dark-brown and black forms are frequent; and 

 the pattern is usually distinct, even if variable. The hind tibiae are 

 frequently red, but this cannot be taken as absolutely diagnostic. 

 During the breeding season males of L. migratoria become a bright 

 yellowish, while males of L. danica show no color change when adult. 



The situation regarding coloration of the hoppers is different. 

 Locusta danica nymphs may be uniformly green, fawTi, gray, brown, 

 or even black. "Quite the opposite is the case with migratoria, in 

 which each larval stage exhibits its constant color characters. Their 

 coloration presents a combination of black and orange-red (or yel- 

 low), the earliest stages being almost entirely black, while orange 

 or yellow appears later and extends with each molt without, how- 

 ever, entirely replacing the black." While there is variation even 

 in this form, Uvarov says: "The main point is that this type of 

 coloration never occurs in hoppers of danica in spite of the wide 

 range of coloration in the latter." Somewhat similar conditions are 

 exhibited in Figure 35 for the solitary and swarming phases of the 

 South African locust, Locustana pardalina, and for the desert locusts, 

 which have previously been known respectively as Schistocerca flavi- 

 ventris and S. gregaria. 



In L. pardalina, of South Africa, there are approximately the 

 same differences between the swarming and solitary phases as have 



