Coxal glands of the Arachnids. 243 



The exit tubules (ET) on the third appendage can be seen in 

 photographs 15, 15 a, 17, 17 a, and on the fifth appendage in 18 

 to 18 c. It has not been found possible to get sections on the fifth 

 appendage showing the saccule, collecting tubule and exit tubule 

 all in one field as is seen so clearly in 15 and 15a for the third 

 appendage. 



In Atijpus, the two saccules and two outlets are present, but 

 the labjTinth appears to be less coiled than in the Theraphosids 

 proper. There is already a tendency toward simplification, but the 

 coxal glands of AUjpus approach much more closely to the Thera- 

 phosid character than to that of the Araneae verae. The finer details, 

 however, could not be determined as I only had at my disposal 

 alcohol specimens kindly provided by M. Eugene Simon. 



It may be remarked that the aberrant IscJmothele, wiiich, unlike 

 the other members of the Theraphosid group, have become web 

 spinners, have lost the saccule and outlet of the fifth appendage, 

 retaining only those of the third appendage. The labyrinth also is 

 greatly reduced and simplified, so that the characters of the coxal 

 glands in this aberrant genus resemble rather those of the Dysderids 

 than of the Theraphosids in general. In the photograph (19) from 

 Ischnothele, the labyrinth tubule leading from the saccule is seen to 

 run posteriorly as a single elongated tubule without coiling, but the 

 walls are irregular in outline, indicating the probability that the 

 straightening out has been caused by the breaking through of ad- 

 jacent coiled tubules. There is evidence also in the other Thera- 

 phosids of this breaking through with consequent fusion of adjacent 

 coils, but it is only in IscJmothele that the process of straightening 

 out has gone far enough to enable one to follow up the course of 

 the tubule. 



It is probable that but few well fixed specimens of the Thera- 

 phosid spiders have been available for studj', and there is some 

 apparent discrepancy in the reported observations. Bektkau (1885) 

 in AUjims found an outlet on the fifth appendage, and Sturany (1891) 

 confirmed this observation, but could find no outlet in a single alcohol 

 specimen of Mtjgale, and Pelseneer (1885) had also previously come 

 to the conclusion that in a large Theraphosid examined by him 

 there is no efferent duct either to the exterior or to any internal 

 organ. 



Schimkevitch (1911) has recently studied the development of 

 a Theraphosid {Isclmocoliis) from Java, and in the embryo found an 



