172 Pi'oceedings of the Royal Physical Society. 



lopodite. At the point where the nerve to the " anvil " 

 leaves the portion taken up by muscle and enters the mass 

 of connective tissue, there is a small ganglion upon it, con- 

 sisting of three or four cells of the ordinary type, surrounded 

 by a connective tissue envelope, and disposed irregularly in 

 the course of the nerve. I shall have more to say of the 

 function of this later on. 



The nerve-endings are surrounded by blood-sinuses, and 

 are slung in these by trabeculae of connective tissue, and the 

 surface of the nerves and nerve-endings is covered with a 

 continuous endothelium, which may be demonstrated by in- 

 jection from the heart with nitrate of silver solution, when 

 the characteristic pattern becomes evident. The nerve-fibres 

 are directly continuous with the tactile end-organs, which lie 

 usually at an acute angle with the surface of the hypodermis, 

 in continuation of the direction of the nerves themselves {see 

 PL VIII., Fig. 3). They vary considerably in length, from 

 O'l mm. to 0'2 mm., and exceed the nerve-fibres slightly 

 in breadth, but not very greatly; that is, they are from '015 

 mm. to '030 mm. The largest are those nearest the tip of 

 the claw where the setse in the tactile tufts are most nume- 

 rous. They are approximately cylindrical; are surrounded 

 by a homogeneous, higlily-refracting membrane, continuous 

 with the trabeculae of the connective tissue which sling the 

 organ, and on whose outer surface are granular nuclei pro- 

 jecting into the blood-space, the same as those found through- 

 out the blood system. The contents of this membrane consist 

 of a very granular protoplasm {i.e., after treatment with re- 

 agents), in which are imbedded nuclei. Both of these ele- 

 ments behave towards re-agents in the same way as those of 

 ganglion-cells, and the apparatus is probably to be regarded 

 as a ganglion-cell formed peripherally. The number of the 

 nuclei varies greatly, the average number being ten or twelve, 

 and their disposition is also undetermined ; they may lie in 

 double or single rows ; this apparently depends on the direc- 

 tion in which the process of nuclear division has taken place. 

 The nuclei are very often found in pairs (as in PL YIIL, 

 Fig. 3), as if they had recently divided, but I have not yet 

 found any trace of a nuclear spindle even in young animals. 



