ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 989 



liuman excrement can easily be seen ; and if it is small, and has been 

 carried a long way by the current, it can usually be recognized by 

 means of the hairs of oats derived mainly from the dropj)ings of 

 horses, which resist decomposition for a long time, and are not con- 

 sumed as food by minute animals. I, however, do not propose to 

 enter into detail in connection with this part of my subject, but 

 specially desire to call attention to the connection between the 

 niunber of minute animals and plants, and the character of the water 

 in which they live, and also to their influence in removing organic 

 impui'ities. 



For some time past I have been carefully ascertaining the number 

 per gallon, of different samples of river and sea water, of the various 

 small animals which are large enough not to pass through a sieve, 

 the meshes of which are about 1/200 part of an inch in diameter. 

 The amount of water used varies from ten gallons downwards, accord- 

 ing to the number present. By the arrangements used there is no 

 important difficulty in carrying out the whole method in a satisfactory 

 manner. I confine my remarks entirely to general mean results. 



The chief animals met with in fresh water are various entomos- 

 traca, rotifera, and the worm-like larvaB of insects. I find that the 

 number per gallon and percentage relationships of these mark, in a 

 most clear manner, changed conditions in the water, the discharge of 

 a certain amount of sewage being indicated by an increase in the 

 total number per gallon, or by an alteration in the relative numbers 

 of the different kinds, or by both. All my remarks apply to the 

 warm part of the year, and not to winter. 



It is known that entomostraca will eat dead animal matter, though 

 probably not entirely dependent on it. I have myself proved that 

 they may be kept alive for many months by feeding them on human 

 excrement, though they soon died without it. If the amount of food 

 in any water is small, not many of such animals can obtain sufficient ; 

 but if it be abundant, they may multiply rapidly, since it is asserted 

 that in one season a single female Cyclops may give rise to no less 

 than four thousand millions of young. In stagnant muddy ponds, 

 where food abounds, I have found an average of 200 per gallon. In 

 the case of fairly pure rivers the total number of free-swimming 

 animals is not more than one per gallon. I, however, found that 

 where what may be called sewage was discharged into such water the 

 number per gallon rose to twenty-seven, and the percentage relation- 

 ships between the different groups of entomostraca were greatly 

 changed. In the Thames at Crossness, at low water, the number was 

 about six per gallon, which fell to three or four at Erith, and was 

 reduced to less than one at Greenhithe. 



There is, however, a very decided limit to the increase of entomos- 

 traca when the water of a river is rendered very impure by the 

 discharge of too much sewage, probably because oxygen is deficient, 

 and free sulphide of hydrogen present. Such water is often charac- 

 terized by the great number of worm-like hirviu of insects. Thus, in 

 the Don, below Sheffield, in summer, I found the number per gallon, 

 of entomostraca only about one-third of what it is in pure waters ; 



