132 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



inner), while the outer parts of the laminae are covered by cells full of secretion. Occasionally, 

 however, non-secreting areas may be found near the outer attached edges of the lamina' as well. 

 No secretion has been found in the cavity of the organ. The sum and substance of these facts 

 probably amounts to this, that the secretory function of the epithelium of the lamina? is exercised 

 periodically only, and that my material was collected during the period of preparation, but 

 before the entire secretory area had assumed its function. 



The epithelium of the lamina 1 appears to be ciliated, and this appearance is connected with 

 the presence of a second sort of cells in the epithelium. I have stained a number of preparations 

 with borax carmine, and then with Lyons blue. Such preparations show a second line of nuclei 

 at the level of the middle of the epithelium of the lamina'. These nuclei are exceedingly slender, 

 being 6/< to S/< in length by In to l\n in width. They belong to and are situated near the 

 center of long thread-like sensory cells, which stand thickly around the secretory cells. (Fig. 73.) 

 The ends of the former project beyond the latter, forming sensory spikes, which are so numerous 

 that thej' give the appearance of a thick coating of strong cilia belonging to the secretory cells. 

 The cell bodies of the sensory cells are so slender and absolutely thread-like that they are not 

 clearly visible without special stains, and then only in places where the other epithelial cells 

 have been accidentally separated. Fig. 73 is an accurate drawing of such a place under a mag- 

 nification of nearly 500 diameters, but taken from the glandular portion of the organ and not 

 from the laminae. The sensory cells, however, are alike in both regions. Favorable cross 

 sections of the epithelia show the nuclei of the sensory cells thickly clustered around and 

 between the secretory cells. The bases of the sensory cells pass into the subepithelial tissue as 

 tine fibres, and are there lost; but it seems onh T reasonable to suppose that they are directly con- 

 tinuous with nerves, and that the cells are true sensory elements. We have now a better ground 

 than before for supposing this organ to have a sensory function, which may very possibly be 

 olfactory. 



The epithelium of the surface of the anterior parts of the organ and of the skin forming 

 the walls of the atrium is of the same form as that of the laminae, though apparently not at 

 all glandular. 



The glandular portion of the organ is composed almost entirely of long branched glandular 

 tubules with narrow lumens. (Figs. 66, 67, and 6S.) The tubules are parallel to each other, while 

 the submucous tissue separating them is so slight in amount as to form little more than a 

 separating lamella. (Fig. 68.) The tubules are lined by a single-layered epithelium, composed, 

 like that of the laminae, of secretory and sensory cells. 



The secretory cells are of a very different character from those of the laminae and probably 

 produce a different secretion. The regular arrangement of their brilliantly staining nuclei in 

 the bases of the cells causes stained sections to appear almost diagrammatic. I do not think 

 that the histological condition of my material is good enough to warrant my making a detailed 

 comparison between these cells and the secretory cells of the laminae. The regular arrangement 

 of the tubules make the glandular region a striking feature of sections. The clearness of its 

 secreting cells compared with those of the laminae, their slightly greater width, and the larger 

 intensely staining nuclei are features which quite clearly distinguish the secreting cells of the 

 glands from those of the laminae. 



The gland cells average 90^u in length and are from 7/j to lO/i in width. The secretion 

 collects in the shape of numerous droplets or granules in the portion of the cell above the nucleus. 

 No secretion was found in the lumens of the glands. 



Around and between the secretory cells are sensory cells exactly similar to those described 

 in the epithelium of the lamina' (Fig. 73). The slender nuclei of the sensory cells form a quite 

 distinct row at the middle of the epithelium. The sensory cells are not so numerous in the 

 glands as upon the laminae, but still their number is surprising. Fig. 73 is drawn from a section 

 in which the secretory cells had separated from the submucosa and each other, leaving the sen- 

 sory cells revealed. The section was cut somewhat obliquely, and so does not show the free ends 

 of the cells. The sensory cells are perhaps a little more numerous here than in most portions of 



