274 bulletin: museum of comparative zoology. 



There is only one arrangement of fibrils which would so nearly imi- 

 tate a network in appearance that the possibility of its occurrence in the 

 visual cells of Limax deserves serious consideration. I refer to a distally 

 branching system of fibrils. From the optic nerve the single fibril or 

 the few fibrils of the neurite might enter the cell body and branch 

 dichotomously, or otherwise, in such a way as not to form a network of 

 any kind. Such an arrangement would be difficult to distinguish from 

 a true network, even in perfect fixation, and especially so if accompanied 

 by some fusion of fibrils. If any one insists that there is a branching 

 system in the visual cells of Limax, the idea cannot be well refuted, 

 for we must rely on the microscopic appearances, which might be very 

 deceptive sometimes, as I have just admitted. But the three methods 

 which I have used in this research check one another ; and since they 

 present practically the same kind of evidence in each case, it is safe to 

 conclude that there is a network in the vicinity of the nucleus. If we 

 admit that an alveolar condition of the cell is the embryonic condition, 

 it is easy to see how a network might arise from an alveolar system by 

 the disappearance of the plane faces and a persistence of the edges of 

 contiguous alveoli. From a physiological point of view it would make 

 little difference whether the fibrils formed a branching system or a net- 

 work, for in either case the cell would function as a sensory unit. 



The most unsatisfactory part of the sensory cell in which to demon- 

 strate fibrils is the neurite. Although on analogical grounds one would 

 expect to find that the neurite consists of at least two or three fibrils, I 

 have been unable to find evidence which would warrant that statement. 

 The neurites of the optic nerve are so small that it is quite possible they 

 may represent single fibrils, as Hesse believes is the case in the neurites 

 from the sensory cells of the eyes of Patella and the cephalopods. But 

 the large number of fibrils in the rod of Limax suggests the likelihood 

 of more than one fibril in the neurite. I have told of the few indications 

 which were found in vom Rath and methylen-blue preparations of a fibril- 

 lated neurite, but the evidence is not sufficient to be conclusive. 



From the methylen-blue preparations there has been no evidence of a 

 union between sensory cells, such as Hilger imagined. All of the evi- 

 dence is in favor of the idea that each rod is represented in the optic 

 nerve by a neurite, which is either one fibril or contains fibrils from its 

 own sensory cell only. This arrangement would of course be favorable 

 to sharpness of vision. 



The discovery of the accessory retina of Limax was a most fortunate 

 thing for the study of the neurofibrillae of gasteropods, for this structure 



