ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 587 
chromocytes containing pigment-granules, and as tlieir combinations 
vary, different colours are observable at different times. In B. violaceus 
there are the same three kinds as in B. smaragdus , but in different pro- 
portions. It is evident that varieties must not be based on differences 
of coloration. The author does not feel inclined at present to admit 
that the movements of the granules are nothing but Brownian move- 
ments. 
Alleged Otocysts of Salpidse.* — Mr. M. M. Metcalf calls attention 
to a blemish in the ‘ Traite de Zoologie concrete,’ by Delage and 
Herouard, a work for which at the same time he has the highest appre- 
ciation. In 1893 Metcalf show r ed that the so-called otocysts of Salpa 
are really glandular organs very possibly related to the neural gland 
of some of the Ascidians. Delage and Herouard have accepted this 
observation, but suggest that the small (accessory) eyes in the ganglion 
of Salpa may be regarded as otocysts. But the cells of which these 
structures are composed are rod-cells exactly like the rod-cells of the 
larger eye. except in size; they have no resemblance to the setigerous 
cells of an otocyst ; there is no cavity near them in which an otolith 
could lie. The idea that Salpa has ears is a zoological myth; for 
there is no trace of any structure which could be interpreted as an ear, 
in the region of the brain, in any of the eleven species of Salpidae 
which Metcalf has studied. 
Regeneration and the Germ-layer Tkeory.f — Dr. L. S. Schultze has 
studied the regeneration of the ganglion of Giona intestinalis , and makes 
this a text for a general discussion of regenerative phenomena. In this 
particular research an interesting point in technique was observed. In 
order to kill the animals in the expanded condition, chrom-acetic acid 
was added drop by drop to the sea-water in which they were living. It 
was found that animals paralysed by this treatment could be revived by 
removal to clean water, and therefore the reagent was used before operat- 
ing as an anaesthetic. In this way the ganglion could be removed with 
much greater ease than in active animals displaying the usual contrac- 
tility. After the operation the animals were allowed to recover in clean 
water. The experiments showed that the ganglion and adjacent struc- 
tures are completely regenerated. This is accomplished by the peri- 
brandhial epithelium, and as this is an ectodermal structure, the new 
ganglion arises from the same layer as that which during embryonic 
development produced the old one. 
In his general discussion of regeneration, the author suggests the 
following definition of a germ-layer : — The germ-layers are complexes 
characterised alike by the organs to which they respectively give rise, 
and by the fact that they respectively occupy definite positions in the 
Metazoan embryo. They are formed of embryonic cells originating 
directly in the segmentation of the ovum. This definition readily covers 
ectoderm and endoderm ; but the conception of mesoderm can only be 
purely topographical. The conception of germ-layer should be limited 
to the embryonic cells arising from the segmentation of the ovum, and 
the layers in budding and regeneration may be designated by the terms 
* Anat. Anzeig., xvi. (1899) pp. 301-2. 
t Jen. Zeitschr. f. Naturwiss., xxxiii. (1899) pp. 263-344 (2 pis.). 
