303 



is greatly thickened and forms a sensory patch consisting essentially 

 of a thin roof of protoplasm formed by the expanded ends of long 

 columnar or flask-shaped supporting cells, in the interspaces between 

 which are situated rounded cells, each with a stiff pointed process 

 which projects through the protoplasmic roof into the lumen of the 

 organ. The nerve enters directly into this sensory patch. Pinkus 

 likens the structure of the organ in Protopterus to that of the ma- 

 culae acusticae. He, however, neither mentions nor figures the "hair 

 cells" with their projecting processes. 



I have also had the opportunity of looking through a series of 

 sections of Protopterus larvae collected by the late Mr. J. S. Budgett 

 and now in possession of Professor Kerr. The development of the 

 organ is precisely the same in this genus, with the exception that the 

 epiblastic ingrowth of the spiracular rudiment is less extensive and 

 well marked. It is, however, easily to be made out. The oldest stage 

 examined was one corresponding with Kerr's stage 36 in Lepidosiren. 



With reference to the development of the spiracle in Ceratodus 

 Kellicott only says that the hyomandibular pouch is never perforated. 

 (Development of the Vascular and Respiratory Systems of Ceratodus. 

 Mem. New York Acad. Sc, Vol. II, Part IV, 1905.) 



This organ has no relation to the lateral line system of sense 

 organs. The early development of these organs presents a totally 

 different appearance. The thickening of the epiblast is in these only 

 two cell-layers thick, and does not appreciably project down below the 

 general level of the ectoderm (Figs. 1 and 2). Pinkus' Organ is in 

 fact unmistakeably formed from the epiblastic portion of the rudiment 

 of the spiracle, the epithelium of which becomes the sensory epithelium 

 of the sense organ. The lumen, which appears about stage 32 is at 

 first a simple slit elongated in the direction of the line joining the 

 skin and pharynx. It never, of course, opens either to the exterior 

 or into the pharynx. 



As the proximity of the epiblastic ingrowth to the facial ganglion 

 in Fig. 1 might suggest that this is one of the branchial sense-organs 

 of Beard it may be mentioned that this proximity is secondary. At 

 an earlier stage (30) the two structures are quite separate, the ganglion 

 being some way behind the rudiment of the sense organ. Although 

 in the entire absence of experimental evidence any suggestion as to 

 the function of this sense organ is of little value, still there are three 

 points in this connection sufficiently interestly to be worth mentioning. 

 Firstly, the organ is so deeply embedded in bone and cartilage as to 

 make it scarcely conceivable that it could receive direct stimuli from 



