392 DISCOVERY REPORTS 



They are present in the large majority of interphase cells, but appear to be absent in cells undergoing 

 mitosis, where the nucleus or spindle lies up near the free end of the cell. The chromatic vacuole 

 consists of blobs or crescents of darkly staining matter deposited irregularly around the walls of a 

 clear vacuole. Fixation and subsequent treatment were compatible with the preservation of lipoid. 

 It is, however, too early to say what relation, if any, these structures bear to the Golgi element. 



The chromatic vacuole has not been identified in the remaining specimens, which were fixed in 

 Zenker, Helly and F.W.A. (without post-osmification). Whether this is because the fixatives were 

 unsuitable, or because the gas-glands were older or inactive at the time of fixation has yet to be 

 determined. Dahlgren and Kepner did not give the size of their specimens, nor did they mention the 

 histological techniques they employed. 



With regard to the secretory activity described by Dahlgren and Kepner, bubbles or vacuoles have 

 from time to time been observed in the distal part of these cells. Such cases are rare in the author's 

 material. Granules have not been seen, and the distal region typically shows a homogeneous or finely 

 fibrillar content, representing normal fixed cytoplasm. However, the histological picture would be 

 expected to vary according to the physiological state of the gland when fixed, and Dahlgren and 

 Kepner's material may have been in a more active secretory state than the author's. The fact that 

 bubbles were found in the Canary material, however infrequently, does constitute evidence in support 

 of Dahlgren and Kepner's general hypothesis. 



Dahlgren and Kepner figured and described bundles of mesogloeal matter interspersed amongst the 

 bases of the gas-gland cells. Such bodies have not been observed in the author's material. 



Groups of columnar cells have been found which, from the size of their nuclei, would appear to be 

 tetra- or octoploids. The cells are in other respects indistinguishable from the remaining columnar 

 cells. This condition has already been encountered in the general ectoderm and endoderm of the 

 float, and no particular significance need be attached to it. It appears to be fortuitous, and the result 

 of metaphase fusion in binucleate cells. 



(b) Giant cells 



In all of the specimens examined, isolated giant cells occur in the gas-gland ectoderm. They 

 occupy a sub-epithelial position, being covered by slanting columnar cells. In young gas-glands they 

 are fairly evenly distributed; their nuclei are rounded and compact, and stain densely in Feulgen 

 preparations (PI. XXVIII, fig. i). In older specimens, the giant cells are separated from one another by 

 wide expanses of columnar epithelium. Evidently, while the latter multiply by regular mitosis, causing 

 expansion of the gas-gland, the giant cells do not divide. Even in the most mature examples studied, 

 the majority of the giant cells occur singly. In the older gas-glands, the giant cells show considerable 

 lobulation and distortion of the nucleus. There is no sign, however, that they are degenerate or pyc- 

 notic. A similar distortion characterizes normal muscle nuclei in young, healthy specimens, and this 

 too increases with age. A large number of nucleoli may be present, and the cells vary considerably 

 in size, the larger ones having the greater numbers of nucleoli. It has not yet been possible to establish 

 clear polyploid groupings on a basis either of nuclear volume or number of nucleoli, but it seems very 

 probable that we are dealing with a case of ' endoploidy ' (' endomitotische Polyploidisierung ', Geitler, 

 x 953)- ^ i s interesting to find that in the salmonoid genus Argentina the gas gland also contains giant 

 cells with deformed, fragmented nuclei (Fange, 1958). 



(c) Islet cells (PI. XXVIII, fig. 2) 



Cells of this type have been found only in one specimen (g = o-S cm.). The specimen was fixed in 

 F.W.A. , which gives a most faithful and delicate fixation, well-suited to the detection of fine cytological 

 detail. The islet cells show up well in iron haematoxylin, thionin and Feulgen preparations. 



