THE ZOOLOGICAL STATION AT NAPLES. 355 



only the free ammonia and the albuminoid ammonia present. The 

 method used is the well-known one of distilling off a portion of the water, 

 and determining the ammonia present in the distillate colorimetrically 

 after the addition of Nessler's reagent, and then further distilling after 

 the addition of an alkaline solution of permanganate of potash, thereby 

 obtaining in the distillate what is known as the albuminoid ammonia 

 present in the water. The values thus obtained afford a fair criterion as 

 to the comparative amounts of organic impurity present. 



As, at the time of writing, I have been engaged on this work only 

 about two months, the results obtained must necessarily be regarded as of 

 a somewhat preliminary nature. Such as they are, however, they seem 

 to show that the method is one of some value. A good many experiments 

 have been made to determine the most favourable conditions for the 

 removal of the ammonia, which is present in large quantities in the tank 

 water of the Aquarium, by means of vegetable life. It has been found 

 that the alga Ulva removes the free ammonia present fairly rapidly, though 

 it has not much influence on the albuminoid ammonia. Indeed, if the 

 amount of this alga taken is moi-e than about one square inch per litre of 

 water this actually increases in amount, presumably owing to the meta- 

 bolism of the vegetable tissues. The alga acts best in diffused daylight, it 

 being bleached in sunlight. Larvae allowed to grow in water purified by 

 this moderate amount of alga were found to be 14 per cent, larger than 

 those grown in the unpurified water. If too much alga has been added 

 they are smaller than the normal, or do not develop at all. 



Probably vegetable life exerts its influence most powerfully through 

 the agency of minute algae and diatoms. Thus it was found that every 

 grain of the coarse sand which is placed in the tanks of the Aquarium is 

 covered with a thin layer of algse and diatoms, and that in water filtered 

 fairly rapidly through a layer of this sand a few inches deep the amount 

 of free ammonia present is diminished to about a fifteenth of its previous 

 amount. 



Several other possible means of purification of the water have been 

 examined. Thus it has been found that if water be exposed to sunlight 

 a few days in a flask filled up to the neck, whereby very little surface 

 comes into contact with the air, the amount of ammonia present is, if 

 anything, increased, but yet larvae subsequently grown in such water are 

 15 per cent, larger than the normal. Larvae grown in water which has 

 been exposed to the action of the air as well as the sun in a flat, covei-ed 

 glass jar are, however, rather smaller than the normal, and also the 

 ammonia present in such water is appreciably increased. Larvae grown 

 in water previously heated to boiling are slightly increased in size. 



Again, a series of observations is being made upon the relative capa- 

 cities for ' fouling ' water possessed by various members of the animal 

 kingdom. Various animals of known weights are placed for known times 

 in measured volumes of water, and the increase in the amount of ammonia 

 present is then determined in a portion of the water. In another portion 

 larvae are allowed to develop, and so by subsequent measurement the 

 adverse effect on tlieir growth of the products of metabolism of the various 

 animals is determined. Still, again, other observations are being made as 

 to the effects upon the growth of the larvae of the addition of various 

 salts, such as nitrates, nitrites, and ammonium salts, to the water. 



In conclusion I wish to offer my thanks to the Committee of the 

 British Association for the privilege of being allowed to hold the table, 



A A 2 



