240 Miscellaneous. 



paper referred to. The polymorphic forms of I. Ajax do not conform 

 to the law of substance-ivaste. Perhaps Mr. Scudder, who has ample 

 opportunities of studying this interesting species, may be induced to 

 turn his attention to the subject and attack it experimentally. 



August 3, 1874. 



Tube-building Amphipoda. By S. I. Smith. 



In examining recently an alcoholic specimen of a species of Xe- 

 noclea, I noticed a peculiar opaque glandular structure tilling a large 

 portion of the third and fourth pairs of thoracic legs, which in most, 

 if not all, the non-tubebuilding Amphipoda are whoUy occupied by 

 muscles. A further examination shows that the terminal sqgment 

 (dactylMs) in these legs is not acute and claw-like, but truncated at 

 the tip and apparently tubular. In this species, a large cylindrical 

 portion of the gland Ues along each side of the long basal segment, 

 and these two portions uniting at the distal end pass through the 

 ischial and along the posterior side of the moral and carpal seg- 

 ments and doubtless connect with the tubular dactylus. There 

 can be no doubt that these are the glands which secrete the cement 

 with which the tubes are built, and that these two pairs of legs 

 are specialized for that purpose. 



A hasty examination revealed a similar structure of the corre- 

 sponding legs in Amphithoe rnaculata, Ptiloclieirus pinguis, Cera- 

 pus rubricoi'nis, Byblis Gaimardi, and a species of Ampelisca. 

 In all these except the last two a very large proportion of the 

 gland is in the basal segment. In the Amphithoe this segment is 

 thickened and the gland is in the middle. In the Cerapus it is 

 very broad and almost entirely filled by the gland, with only very 

 slender muscles through the middle, and the orifice in the dactylus 

 is not at the very tip, but subterminal on the posterior side. In 

 the Ptilocheirus the gland forms three longitudinal masses in the 

 basal segment and is also largely developed in the meral and car- 

 pal segments. The dactylus is long and slender, and the orifice 

 subterminal. In Ampelisca and Byblis (which, like Haplo'ops, 

 are tube-building genera) the meral segments of the specialized 

 legs are nearly as large as the basal, and contain a proportionally 

 large part of the gland. In these genera the remarkable elonga- 

 tion of the two distal segments in the third and fourth pairs of 

 legs is perhaps a special adaptation to enable them to reach back 

 over the deep epimera. 



The examination of fresh specimens will doubtless show these 

 structures much more fully. — SiUiman^s American Journal, June 

 1874. 



