326 SPENCER WALPOr,E AND T. H. HUXLEY. 



skin, or epidermis, covrring a deep fibrous and vascular true 

 skin, or derma. The former is divisible into a superficial, a 

 middle, and a deep layer of cells, the last being in immediate 

 contact with the derma. The deep cells are vertically elon- 

 gated, the middle ones more or less broadly spindle-sha[)ed 

 or rounded, while the thin superficial layer consists of flat- 

 tened cells. The deep cells are constantly multiplying by 

 fission, and their progeny become middle cells, the outer- 

 most of which, for the most part, becoming flattened, give 

 rise to the superficial layer, which is continually shed and 

 replaced. Some of the cells of the middle layer, however, 

 enlarge, take on a more or less spheroidal form, and become 

 filled with a mucous fluid. As they rise to the surface, they 

 open and pour out this fluid, which lubricates the surface of 

 the fish. In any vertical section of a properly prepared 

 portion of salmon skin more or fewer of the openings of these 

 cells are to be seen. The derma is composed of matted 

 bundles of connective tissue, traversed by blood-vessels and 

 nerves, and containing numerous lymphatic spaces. The 

 superficial layer of the derma contains a number of dark 

 pigment cells, of which there is a close-set zone immediately 

 beneath the epidermis. 



In a thin vertical section of the skin of the head of a 

 salmon, which has passed from the sound skin through the 

 centre of a diseased patch, the various structural elements 

 which have been described, disposed Avith great regularity, 

 are alone visible in the healthy part of the section. But, on 

 advancing within the margin of the diseased area, hyphse of 

 the Saprolegnia are seen to penetrate horizontally between 

 the cells of the middle layer, thrusting them asunder with 

 so much force that the cells become bent and distorted, and 

 adhere to a hypha as if they were spitted on it. And, in 

 fact, it is because bundles of such hyphae are thrusting them- 

 selves in this manner, as the roots of an ordinary plant 

 thrust themselves into the soil, between the epidermic cells, 

 that the radiating ridges Avhich appear on the marginal area 

 of the diseased patch are formed. Close up to the free ends 

 of these hyphse, however, the epidermis is perfectly healthy ; 

 and this fact suffices to prove that the growth of the fungus 

 is the cause of the morbid affection of the epidermis, and 

 not its consequence. 



Proceeding further towards the centre of the diseased 

 patch, the hyphse become more numerous and take a vertical 

 as well as a horizontal direction. Of the vertical ones^ some 

 traverse the epidermis outwards, thrusting aside and disturb- 

 ing its cells^ and terminating in short free ends on the 



