FISHES. 369 



the oxygen into carbonic acid, in making it pass through their intes- 

 tines. Such is the cobites, according to the curious experiments of 



(lition in which fishes are found in their natural state, and on removing them from 

 the water in which they live, we may regard them as saturated with this liquid, pro- 

 vided they are in a state to absorb it. Now we shall take the measure of their ca- 

 pacity of saturation with water, as we have hitherto done in regard to the batra- 

 chians, the loss which they experience by perspiration before death ; and we see it 

 is sufficient to ensure the death of fishes, that they lose the fourteenth or fifteenth 

 part of their weiarht. If this loss appear too inconsiderable for us to ascribe the 

 death of these animals to it, let us compare this result with those which we ob- 

 tained in our researches on the batrachians. They were not given in the preceding 

 chapter, that they might be reserved for this occasion. It has been shewn that the 

 point of saturation with wat?r, in the case of the batrachians, depends on the state 

 of their nutrition, and that it may vary within very considerable limits. Now, the 

 losses which they undergo by perspiration vary in the same manner. In conditions 

 favourable to nutrition, their capacity of saturation may equal the third of their 

 weight; but, in unfavourable conditions, it is so small, that the least appreciable loss 

 is suflicient to cause death. On applying these results to fishes, whose capacity for 

 water is small compared to that of batrachians, we shall see that the loss which they 

 experience by evaporation is enough to cause their death in air. But the pheno- 

 mena relative to this subject are not always so simple ; they may be very compli- 

 cated : one might be led to believe, that atmospheric respiration would keep fishes 

 alive, if we could devise means for obviating their loss of weight by evaporation. 

 With this view, a fish which had been wiped and then weighed, was suspended in 

 a limited quantity of aerated water, so that it had its iiead and gills above the 

 surface ; it died in nine hours and twenty-one minutes. On then weighing it again, 

 it appears that it had not sensibly diminisiicd in weight, but on the contrary, had 

 slightly increased. This result would appear to be independent ol the cause we 

 have before assigned for the death of fishes, where the whole body is exposed to the 

 action of the atmosphere. Hut before enquiring into the influence of a new cause 

 which may be added to the first, let us more attentively examine the complicated 

 case in which fishes are found in the circumstances of the experiment last related. 

 The body is phinged in water, but the head and gills are exposed. On one hand ab- 

 sorption takes place in the water, ou the otlier, perspiration in the air. The ab- 

 sorption by the body plunged in water is proved by the slight increase of weiglit 

 which takes place during the experiment, and the loss by perspiration from the part 

 exposed to the air is demonstrated by the preceding experiments. Now it is evi- 

 dent, that the organ of respiration, which is exposed to the atmosphere, cannot 

 continue its functions unless the losses by perspiration are repaired. It is true, 

 the rest of the body absorbs, and that on tlie whole, it does not lose any of its 

 weight; but this condition is not sufficient for the continuance of respiration. It 

 is also necessary that the distribution of the fluid absorbed by the trunk, should be 

 such, that the giUs and muscles which move them should receive a proportion of it, 

 capable of repairing the loss which those organs experience by perspiration. Pre- 

 suming it possible that this e(|uilibrium might not take place, I made the following- 

 experiment to enquire into the relations of partial and simultaneous perspiration 

 and absorption. I placed some fishes in the opposite position to that of the fish em- 

 ployed in the last experiment, that is. with the head and gills in water of the same 

 quality and quantity, and the trunk, suspended in the air by a thread passed through 

 the end of the tail. They lived in this state many days. I weighed them after 

 that interval, and discovered that there was evidently, in this case, a slight increase 

 of weight. JJut the drying of the surface of the part of the trunk exposed to the 

 air, was as marked as in the case where these animals were entirely exposed to the 

 atmosphere, and where they died after a considerable diminution in w-eight. It is 

 therefore evident, that the fluid absorbed by the gills was not distributed to the rest 

 of the body in a proportion sutHcient to repair, in all parts of the trunk, the loss 

 which it had sustained by perspiration in air. 



" The following fact, relative to the physical conditions of fishes in air, is impor- 

 tant in the consideration of the principal causes of their death when so placed. 



VOL. II. B B 



