52 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 



As regards microscopic structure, the state of preservation leaves so much to 

 be desired that it seeins hardly worth while to go into very minute details. 

 The following is an account of the facts I have been able to elucidate by a 

 study of serial sections. 



Each organ consists of the following parts : — 



1. The Capsule. 



2. The Posterior Cup. 



3. The Inner Cup. 



4. The Central Mass. 



5. The Anterior Cap. 



1. TJie Capsule (Plate 9, Fig. 2, c) is about 0.04 ram. thick, and consists of 

 rather loose connective tissue with small irregularly-shaped nuclei scattered in 

 it here and there. The inner half of the thickness of this layer is deeply 

 pigmented, the tissue being permeated by a mass of granular reddish-brown 

 particles. The color and density of this layer agree with that to be noticed 

 subsequently in other organs of this species, and it is in marked contrast with 

 the very opaque black pigment seen in Abralio2)sis. It is specially noteworthy 

 that no similar layer was observed in the case of Pyroteidhis manjaritifera. 



2. Tlie Posterior Cup (Fig. 2, p. c.) is built up of a large number of flattened 

 oval or almost circular scales arranged in concentric layers, and leaving a 

 hollow in the middle which contains the knob of the central mass. In section 

 these scales, being thicker in the middle than at the edges, have an irregularly 

 fusiform outline, and their concentric arrangement is very conspicuous. They 

 stain very deeply with all the reagents yet tried (haematoxylin, oirmine, and 

 osmic acid). In some parts of the sections, notably in the more superficial 

 layers, they appear to be homogeneous, bat in other parts they present a 

 frayed-out appearance, and are then seen to be made up of a number of inter- 

 lacing fibrils with a less dense substance between them (Fig. 3). They vary 

 much in dimensions; the largest, situated in the outer layers, measure 0.08 mm. 

 in diameter by 0.03 in thii^kness, whilst those near the centre measure rather 

 less than half these dimensions. The smallest of all are near the rim of the 

 cup ; they are sub-circular rather than fusiform in outline, and the diameter in 

 many ca.ses is only about 0.015 mm. 



The scales are not in contact with each other, but are separated by inter- 

 spaces of about O.OI mm. in thickness, but varying much in different jilaces. ■ 

 A delicate connective tissue, which stains only very feel)ly with the reagents 

 above mentioned, fills up these interspaces. Here and there a radially directed 

 pa.ssage is seen perforating several of tlie layers of scales, ami in its centre may 

 be Been a fibril which can often be traced to the central lujiss, and is presum- 

 ably a nerve (Fig. 2, n). I have not, however, been able to trace these nerves 

 to their source in the outer layers of the organ. 



■ The actual form of the pt-riphery of these scales is very dillieult to determine, 

 becau.se the sections cut them at such varying angles. Tangential sections 



