Miscellaneous. 290 



thereof, as supposed by Agassiz and also by Baudelot ; or else these 

 parts are developed at the expense of a special blastema as a kind 

 of teeth, the opinion maintained by M. Mandl. Although this latter 

 view is generally abandoned, the following observations show that 

 it may be regarded as correct, at least in certain fishes. 



Gobias niger (Linn.), very common on our coasts, has fur- 

 nished me with the materials for these researches. Its scales, of 

 which M. Baudelot has given a very good description, are of a very 

 simple type, never having more than a single row of spinules on the 

 posterior margin of the concentrically ribbed lamella. This lamella 

 may even exist alone, either all over the body, as in very young 

 individuals, or onlj r on the ventral parts, as in the adult. 



In the complete scales the lamella shows a nearly quadrilateral 

 form, the anterior side being rounded, and the posterior side slightly 

 projecting and angular. The number of marginal notches seems 

 never to exceed nine or ten. The focus, which is usually small and 

 circular, is marginal, approaching the posterior margin ; the con- 

 centric crests are numerous. The epithelium, in which the pigmentary 

 layer is distinct, clothes all the posterior part of the scale and forms 

 a festoon-like marginal mass, in which the spinules are entirely 

 imbedded. The latter appear in an order well described by authors — 

 that is to say, commencing with a median spinule, then lateral pairs ; 

 at least this is what anatomical examination leads us to suppose, 

 which always shows the spinules unpaired in number when there 

 are not more than seven ; beyond this it is not rare to find an even 

 number, which may no doubt be attributed to the unequally rapid 

 development of the lateral spinules. The observations related here 

 may explain these facts. The length of the spinules differs at the 

 centre and at the sides, of course only taking into consideration the 

 completely developed spinules. The former are sensibly shorter, and 

 the size regularly increases to the spinules nearest the angles ; so 

 that all the points, without coming to the same straight line, form 

 an angle a little more open than that of the posterior margin of the 

 lamella ; only the point of the spinules projects slightly beyond 

 the epithelium, which shows at this point particularly well the 

 pigmentary masses distributed in two layers — the first bordering the 

 lamella at the base of the spinules, and the second formed of usually 

 isolated chromoblasts placed in the interspinular spaces all at the 

 same level, producing a very elegant design. 



The epithelial cells, it is very important to notice, are of extreme 

 delicacy ; simple contact with the sea-water, even only for ten 

 minutes, swells them, breaks them up, and transforms the whole 

 into a magma in which it becomes impossible to detect the relations 

 of the parts. To repeat the observations it is absolutely necessary 

 to have a fish not only fresh but living ; in this respect Gobius niger 

 is a particularly favourable species, as it may be preserved several 

 days with very little trouble in a very small quantity of sea-water. 

 As soon as the scale is taken off it ought to be placed quickly under 

 the microscope to examine it at once ; if it be desired to continue 

 the observation, the tissues must be hardened immediately by the 



