44 ME. M. LAUEIE ON THE 



the coelomic cavity in Limulus, Scorpio, PTialangium, and Spiders, 

 and the structure in the adults is much the same in all these forms. 

 Bernard * again suggests setiparous sacs as the origin of the coxal 

 glands, but I do not think he can have understood the significance 

 of vrhat has been described in their development. Setiparous sacs, 

 partly developed from the mesoderm and opening freely into the 

 coelom, do not commend themselves to one as morphological pro- 

 babilities. The differences in the various forms have been so fully 

 treated of by Sturany t that it seems unnecessary to recapitulate 

 the details here. The one point on which I wish to lay some 

 stress is the diiference which exists as to the segment to which 

 the coxal gland belongs. In the Scorpion and Limulus it opens 

 at the base of the filth pair of appendages. Kowalevsky and 

 Schulgin X describe it as belonging to the third in Androctonus 

 ornatus ; but my sections of Euscorpius italicus and Centrurus 

 leave no possibility of doubt that in these forms it is the fifth, and 

 as they themselves seem not very certain, I think it probable that 

 they were mistaken. In Phalangium, JPhrynus {supra), and 

 Spiders this organ opens at the base of the third pair of append- 

 ages. Bertkau § says he has seen ducts to the fifth pair of 

 appendages in Atypus, and Sturany says also that the gland opens 

 on the fifth in the Tetrapneumones. In Pseudoscorpions it 

 opens on the " third leg " — which, I presume, means the fifth 

 pair of appendages — according to Bernard. 



It seems, then, that while the coxal glands are serially homo- 

 logous in different forms, they belong to difi'erent segments, and 

 by this character alone the Arachnida would be divided into two 

 sections — one containing the Scorpions and Limulus, in which the 

 gland opens on the fifth appendages, and the other the rest of the 

 group, in which there is a gland in the third segment, with the 

 possible exception of some Spiders and Pseudoscorpions. A 

 gland may also be present in the fifth segment in these forms. The 

 antennary and shell-glands of Crustacea are no doubt structures of 

 the same kind but belonging to different segments, i. e. either the 

 second and fifth or first and fourth, according as one does or does 

 not count the first antennae as somatic appendages. Consequently, 



* Ann, & Mag. vol. xii. 1893. 



t Arb. Zool. Inst. Wien, vol. ix. 



J Biol. Centralbl. vi. 



§ Arch. mikr. Anat. vol, xxiv,, and Zool. Anz. xcii. 



