NO. 8 INSECT ABDOMEN SNODGRASS I 5 



roach become horizontal and the ovary assumes its definitive form 

 and position. In many insects, however, as in Ephemerida, Dermap- 

 tera, Plecoptera, Phasmidae, and Acrididae, the ovarioles preserve 

 their more primitive serial arrangement on the elongate calyx, as at 

 an immature stage of the cockroach (G). 



The development of the testes does not dififer materially from that 

 of the ovaries. The sperm tubes usually lack terminal filaments, and 

 the division of each testis into compartments is not always apparent 

 externally, and is sometimes incomplete. In the Lepidoptera, accord- 

 ing to the studies of Zick (1911) on Picris, the lumen of the testis 

 is undivided until about the time of hatching, when there begins the 

 formation of the septa which will divide the organ into four tubular 

 sections. The septa are in foldings of the testicular wall, which grow 

 inward toward the posterior ventral part of the organ where the duct 

 arises. 



The lateral ducts. — The primitive mesodermal exit tubes of the 

 gonads become the lateral oviducts in the female and the vasa defer- 

 entia in the male, except in so far as they may be partially replaced 

 by branches of the ectodermal median duct. Since the mesodermal 

 ducts are derived from the coelomic walls, they probably originate 

 as channels between epithelial folds continuous with the epithelial 

 covering of the germ cells, which eventually close to form tubes. The 

 idea that the reproductive ducts of Arthropoda are modified nephridial 

 tubes is purely theoretical, and there is nothing to suggest that they 

 are not, from the beginning, canals of the mesoderm formed spe- 

 cifically for the conduction of the germ cells to the outside of the body. 

 Such ducts exist in some of the Annelida, and they may even unite 

 distally in a common outlet tube (Hirudinea). 



The primary mesodermal oviducts of insects undoubtedly opened 

 to the exterior on the seventh abdominal segment, since in embryos 

 and in young nymphal and larval stages of many modern insects they 

 are attached to the ectoderm at the posterior end of the venter of the 

 seventh segment. Heymons and Wheeler were the first to call attention 

 to this condition in the more generalized insects, and the same thing 

 has since been observed in representatives of most of the higher orders 

 by various subsequent investigators. The subject has been so well re- 

 viewed by Nel (1930) that nothing further can be added here to 

 substantiate the general conclusion. Nel quotes authorities to show 

 that the termination of the mesodermal oviducts in the seventh seg- 

 ment is now known to occur in Thysanura, Ephemerida, Odonata. 

 Dermaptera, Orthoptera, Hemiptera, Coleoptera, Lepidoptera, Hy- 

 menoptera, and Diptera. Though some writers say that the openings 



