ARTHUR DEWrXT WHEDON 425 



in size and position that the question of their origin remains 

 obscure. The most marked changes are: 



1. Reduction in the number of muscles. 



2. Reduction in the size of nearly all nmscles. 



3. The development of new muscles connected with the repro- 

 ductive organs. These must develop during or after metamor- 

 phosis as full grown larvae show no distinct beginnings. 



Abdominal ^Modifications as Adaptations 

 When the anatomical conditions described above are gone 

 over in an effort to relate them to the functions of the abdomen 

 it is quite clear that the structures of the larva are direct adapta- 

 tions to the primary functions. In form the abdomen is adapted 

 to wriggling (Zygoptera) or darting (Anisoptera) through the 

 water; the large digestive tract with its extensive dilations and 

 gizzard armature is very well fitted to the feeding habits of a 

 voracious larva; the rectal gills of Anisoptera together with the 

 extensive tracheal system permeating the whole body is un- 

 doubtedly a mechanism developed to meet the respiratory needs 

 of active acjuatic lai'vae; and all the nmscle systems are directly 

 i-elatcd to certain evident functions.* The muscular differences 

 between the Zygoptera and the Anisoptera are perhaps greater 

 than any others, but it would seem that the tubular arrangement 

 and longitudinal direction of the muscles in the Zygopterous 

 larva were perfected for locomotion by wriggling from side to 

 side, and this is practically the only means of swimming they 

 possess. The development of the dorso-ventral muscles in in- 

 creasing degree becomes evident in the Anisoptera, where the 

 larvae are dorso-ventrally compressed, and here the spacious 

 rectal chamber or branchial l)asket is frequently filled with water 

 and suddenly and vigorously emptied, the force of the ejected 

 water being Used for locomotion l)y darting. 



In the imagoes many of the structures are clearly fitted to 

 particular uses, though thej' arc cjuite changed from the status 

 of the larva. However, certain of the conditions following trans- 

 formation are not so clear, and of these the most important are 

 the great elongation of the abdomen, its variation from a cylin- 

 drical form in the Zygoptera to flattened in the Lib(>llulinae, the 



TRANS. AM. EXT. SOC, XLIV. 



