COMPAEATIYE MOEPHOLOGT OF THE GALEODID^. STO" 



then measure 3x2/u. At their largest they are ahout 3^X3 a*, and the chromatin 

 is then more granular. They appear never to hecome swollen and clear like the large 

 nuclei of the main duct. 



These end-tubules are perhaps the homologue of the branching tubes from the proximal end of the 

 coxal gland of Scorpio. In both cases they are supplied with blood by a special arrangement of the 

 parts. It is not easy to ascertain their morphological significance. Are they simply secondary 

 developments of the proximal end of the main duct, or are they primitive coelomic structures comparable 

 •with the end-saccules of the anteunal and shell glands of the Crustacea? These are questions which 

 cannot be answered without further investigation. I am myself disposed to regard them as 

 secondary developments of the proximal ends of the ducts. In each case the epithelium in a way 

 repeats the character of that of the main duct. In Scorpio the cells forming the main duct are large 

 and flat, in Galeodes they are thin and cylindrical. Similarly, in the end-tubules of Scorpio the cells 

 are squamous, with protuberances containing the nuclei ; in Galeodes the cells of the end-tubules are 

 crowded and more or less cylindrical. In Scorpio, further, the epithelium cau be seen gradually passing 

 from the striated form of the main duct into the non-striated form of the end-tubules. Further, the 

 duct of the coxal glands in Obisium appears to me to end blindly and to have no terminal differentiated 

 portion. Sturany found no internal opening in that of a very young Tarantula, and Lomau (50) did 

 not succeed in demonstrating an internal aperture to the gland in the Phalangids. 



On the other hand, Laurie claims to have found the gland opening into the ccelom iu a Scorpion 

 embryo. Until this is confirmed, however, the bulk of e\'idence seems to point to the coxal gland as a 

 blindly-ending tube, with, in certain cases, a branched differentiated portion, which somewhat resembles 

 the end-saccules of certain Crustacean glands. 



»* 



The external aperture of the coxal gland is not easy to find, heing obscvired by the 

 folds of the soft skin behind the first pair of legs. On an excised piece of skin, however, 

 the aperture can be seen as a round hole with thickened edges. In such a preparation 

 from a Galeodes, a curious knob-like structm-e developed from the wall of the duct 

 protrudes through the aperture. 



Uomology of the Coxal Glands. — What is the homology of these coxal glands which 

 occur in nearly all the Arachnids on either the 3rd or the 5th segment of the cephalo- 

 thorax, but never on the ith segment or in the abdomen ? Many investigators, relying 

 chiefly upon embryological evidence, claim them as true nephridia, inherited from the 

 Annelidan ancestors of the Ai-achnida. 



For my own part, I believe that the Arachnids are to be derived directly from Annelids, 

 i. e. without the intervention of any specialized intermediate form. If so, where are 

 the primitive nephridia ? It is no doul)t possible that the coxal glands in the cephalo- 

 thorax are persistent nephridia, and very probable the ducts of the genital glands in the 

 abdomen are adapted nephridial ducts. On the other hand, iu the Annelida, the 

 nephridia tend to disappear from the anterior segments of the body as the posterior 

 segments develop, especially when the anterior region shows great structural modification. 

 We should therefore hardly expect nephridia to persist in the cephalot borax of the 

 Arachnids. My own suggestion that these coxal glands might be developments of the 

 setiparous sacs and therefore homologous with the tracheae on the other segments, wliich 

 would explain (1) their appearing in Line with the tracheae and spinning-glands, (2) their 



50* 



