COMPAEATIVE MORPHOLOGY OF THE GALEODID.^. 371 



liind-gut and are eitlicr lost in vast numbers or else pass out into the body through the wall of the hind- 

 gut. It appears to me highly improbable that these active, living, and, to the last moment, digesting 

 cells can be lost in such numbers, and even more unlikely that they again become epithelial cells after 

 travelling into the hind-gut. 



This detachment of digesting-cells appears to be very general among Arachnids, altliough I have 

 found no cells in the stercoral pockets either in Galeodes or the Spiders. Whether in these animals 

 also they become blood-cells or, after digesting their food contents, are themselves rejected with the 

 faeces, is a matter deserving investigation. In Thehjphonus, I have noticed large, clear, nucleated cells 

 in the blood-plasm in the heart, which are much larger than the ordinary blood-ccUs ; perhaps these 

 were mid-gut cells escaped through the wall of the hind-gut. 



X. PtESPIRATORY SYSTEM. 



ResiDiratiou in the Galeodicliie is effected by means of tracheal tubes ramifying among 

 the tissues. These open to the exterior through three pairs of stigmata, and, in some 

 cases, through an extra median stigma (PI. XXX. fig. 13). 



The Stigmata. — The first pair open behind the coxae of the second pair of legs (on the 

 4th segment). It is protected by an elaboi-ate arrangement of setfe forming tree- and 

 bush-like fringes, which allow the air to pass in while barring the entrance against foreign 

 particles (PI. XXX. fig. 18). Some of the flat-headed setse shown in tliis figure within 

 the ajDerture, appeal- to open at their tips, and may be sensory, i. e. olfactory. 



The 2nd pair of stigmata open slantingly on the 2ud abdominal segment, either under 

 folds of the integument, or, in a few cases, on the surface of the body, and near the 

 median line, while the 3rd pair open still nearer the median line. Tliese are simple 

 chitinous slits (PL XXX. fig. 19), being protected under folds of the skin. In the genus 

 Galeodes the posterior edges of these folds are furnished with remarkable rows of 

 bristles, the " stigmatic combs " (PI. XXIX. fig. 10, PI. XXX. fig. 16, co). These,, 

 perhaps, serve to keep a passage open for the air when the abdominal segments are 

 telescoped together. 



On the 4th abdominal segment in some Galeodidse there is a single median stigma, 

 which often appears to be altogether closed (see the figures). It seems to l^e quite 

 absent in the genus Bhax. 



The position of these abdominal stigmata can best be explained on the assumjitiou that 

 they were apertures on the posterior faces of the coxae of tlie limbs. If the limbs 

 of the 2nd and following abdominal segments folded back in the median line, like the 

 genital appendages themselves (PI. XXIX. fig. 11), the stigmata would be brouo-ht 

 together. This would also explain their slanting positions under folds of the integument. 



The fact that, in the Galeodidfe, we actually have stigmata in the act of aborting" 

 on segment iv., associated with the remains of limbs closely resembling those vesti""es 

 of limbs which function as stigmatic opercula on segments ii. and iii., leads almost 

 inevitably to the conclusion that stigmata have already disappeared from between the 

 similar rudiments of limbs found on segments v. and vi., and perhaps also from those 

 still further back. That there were stigmata at least as far back as the vith abdominal 

 segment in Galeodes may also be inferred from tiie presence of six pairs of ostia in tlxe 

 abdominal portion of the heart, and six pairs of dorso-ventral muscles. These latter, as 



49* 



