238 DR. W. N. F. WOODLAND OX THE 



The metliod by which I achieved success was that of attaching 

 weights to some Perch in a tank in the manner described by Moreau *. 

 The fish, if appropriately weighted, increase the volume of gas in 

 their bladders in order to counteract the sinking action of the 

 weights, but, according to my experience, it is all-important that 

 the weights should not be too heavy, since if the fish are perma- 

 nently attached to the bottom — if there is no hope of being able to 

 rise — they apparently make no attempt at gas-production. The 

 form of weight I employed was a half -hoop of lead weighing about 

 5 gm. (appi'oximately one-tenth the weight of the fish, and this 

 was probably too heavy) ; this I suspended round the lower half of 

 the fish (by thin wiies tied dorsally) in the manner of a belt 

 passing dorso-ventrally between the pectoral and pelvic fins, so 

 leaving these perfectly free. After some twelve hours or moie 

 the fish can swim about more easily and with a lighter weight it 

 would regain all its usual mobility. If now the weight be removed 

 the fish williise quite helplessly to the surface of the water, owing 

 to the increased quantity of gas in the bladder. Unless the fish 

 so rises after the experiment it is of little use to examine the gas 

 gland. It is also, of course, necessary to compare the gas gland 

 sections of the fish experimented on with similar sections of a 

 control fish in the same tank. I weighted some six or seven Perch 

 for 17 hours and at the end of that period the bladder of each was 

 appreciably swollen, the fish having to swim vigorously in oi'der to 

 keep below the surface after removal of the weight. I emptied 

 the bladder in the usual manner by a small puncture posteriorly 

 and in several cases refilled it with Zenker's Fluid ; in others I 

 used a mixture of one volume of Glacial Acetic with tv/o volumes 

 of Absolute Alcohol, and in others 1 "/^ Osmic Acid. Bladders 

 fixed with Zenker I stained with Ehrlich and Picro-indigo-car- 

 mine ; bladders fixed with the Glacial-Absolute mixture I stained 

 with Iron-ha^matoxylin in order to study the details of nuclear 

 structure ; bladders fixed with Osmic I stained with Ranvier's 

 Picro-carmine. In all cases I made surface- view preparations, 

 laying the bladder with its gas gland inside uppermost on the 

 slide, in addition to cutting numerous sections of various thick- 

 nesses of the Zenker and Glacial-Absolute material. The Zenker 

 mateiial gave the best results from the gas bubble standpoint, 

 probably owing to the absence of strong diflusion-currents during 

 the process of fixation ; the Glacial- Absolute mixture, on the other 

 hand, though perfect in its fixation of most parts of the cell, was 

 doubtless somewhat drastic in its action on delicate structures like 

 extruded gas bubbles. 



The sections of the Zenker material showed up the production 

 of the gas bubbles in a very effective manner, the bubbles being 

 clearly seen to arise in the cell-substance (not shown in text-fig. 61, 

 p. 240) and to be extruded as foam-like masses all over the surface 

 of the gas gland lining the bladder (PL IX. fig. 70 and text-fig. 61, 

 the latter being a microphotograph). It must be remarked that 

 only occasional cells were thus active, the majority not exhibiting 

 * A more simple method, of course, is to empty the bladder with a trocai-. 



