of the Salmon in Fresh-Water. 107 



in November fish being about half that in May fish. Second, there it 

 an enormous difference between the average amount of fat in fish at the 

 mouth and the average in those in the upper water during the later 

 months of the year. 



The exact nature of this change in the amount of fat can only be 

 satisfactorily understood when the results of chemical examination are 

 supported by histological evidences. 



1. METHODS. 



Pieces of the lateral trunk muscle both thick and thin, i.e. dorsal and 

 abdominal, from each fish, were fixed in (a) sat. sol. of corrosive subli- 

 mate, and (b) in 1 per cent, solution of osmic acid for 24 hours, then 

 thoroughly washed, hardened in alcohol in the usual way, and quickly 

 dehydrated. They were embedded in paraffin, and longitudinal as well as 

 transverse sections of to 8 microns thickness were cut. Various solvents 

 of paraffin, e.g. cedar oil, xylol, clove oil, benzole, were tried before 

 embedding, and xylol was found to % answer best in preserving the fat. 

 The sections were mounted on albuminised slides to afford every facility 

 for the preservation of fat. It was found by experiments that fixing 

 tissues in reagents other than osmic acid, and afterwards treating the 

 sections with osmic acid, did not give satisfactory results. Sections fixed 

 in corrosive sublimate were stained with some suitable double stain such 

 as hyernatoxylin and eosin or methyl-blue and eosin. The fat in these 

 sections was, to a great extent, washed out, and they then served for 

 comparison with the osmic acid sections, and helped in the study of the 

 general change in the muscles. 



'Ihe reduction of osmium in sections fixed in osmic acid revealed tin- 

 presence of fat globules in a most perfect manner, and it only needed 

 some single stain, e.g. eosin, to bring out other details. Tissues fixed in 

 osmic acid are rather difficult to stain, but eosin seemed to answer the 

 purpose well. 



II. RESULTS OF EXAMINATION OF THE SECTIONS. 

 A. THOSE FIXED IN PERCHLORIDE OF MERCURY. 



1. M-ttscle Fibres. The size of the muscle fibres varied very much, 

 from 50 to 167 microns in diameter, the largest size being in early 

 fish from the mouth of the river. In two specimens caught at the 

 mouth in August, the fibres were about 134 microns in diameter. In 

 those from the upper reaches the size of the fibres varied from 50 to 

 100 microns in diameter, the smallest size being in the thin or 

 abdominal muscle of specimen No. 63, a fish caught in October. 



'2. Mriation. In longitudinal sections the transverse striatkms were 

 well seen in fibres in the centre of a bundle, whereas the fibres at the 

 periphery showed, in many cases, a tendency to longitudinal cleavage, 

 and there the transverse striation was obscure. This longitudinal 

 cleavage was most pronounced in fish from the mouth of the river. 



3. Muscle Vibrillce. The size of the iudividul fibrils in muscle fibres 

 seemed fairly constant in all specimens, being of 1-5 microns to 1'8 

 microns in. diameter. The number of Conheirn's ureas varied with the 

 size of the fibres. 



4. _Nnd<',nu. The longitudinal sections showed well-marked, rather 

 elongated nuclei. No change was noticed in the different specimens 

 except that in No. 79 they were very prominent, but there w.is no 

 Indication of any active proliferation having taken place. 



5. Amount of Fat. The amount of fat in specimens fixed in corrosive 

 sublimate 1 could only bo approximately ascertained by the honeycomb 

 like appearance of tlio emptv spaces in the connective tissue left by the 

 t'at cells, and by a comparison with the osmic acid preparations. In the 



iwH this layer of empty spaces shows an average of hmn. in 



