The Relations between Marine Animai and Vegetable Life. 



359 



Number 



of 



Expt, 



Nature of sanil nsed 



Time of 

 fìltratioii 

 of 1 litre 



»/o Variation in 

 amount of 



freeNHaj -jf^- 



Vo 

 Larvas 

 formed 



Variation 



in Size of 



larvae. 



55 



63 



64 



65 



99 



100 



101 



123 



124 



125 



126 



Coarse Aquarium sand 30' 0" 



Same sand kept stagnating lOdays 22' 0' 

 Same sand previouslyheatedto 70° C.i 30' 0" 



20' 0" 

 3' 15" 

 3' 30" 

 l'45" 

 4' 0" 

 1 h 20' 

 8' 15" 

 2' 40" 



Fine Aquarium sand 



Stagnant Aquarium sand 



Surface shore sand 



Shore sand 1 metre below surface . 



Aquarium sand 



Same sand 



Shore sand 1 metre below surface . 

 Same sand with top crust removed 



— 94 



— 79 



+ 18 



— 13 



— 56 

 + 185 



— 12 



— 87 

 + 358 



— 87 



— 89 



— 37 

 + 272 



— 30 



— 10 



— 4 



— 6 



— 5 

 + 12 



— 6 



— 6 



100 



100 



80 



98 



54 



8 



58 



72 



88 



100 



+ 6.2 

 + 5.1 



(+ 1-3] 

 + 5.4 



+ .6 



+ 4.7 

 (+1.8) 

 + .7 

 + 2.0 

 + 8.5 

 +4.2 



In every experiment given iu this table it will be seen that the 

 filtration of the water through sand had a favourable influence on 

 the growth of the larvai, and also that in the great majority of cases 

 the percentage of ova which developed to eight days Plutei was 

 greater than the normal (i. e. 66.6^). To discuss individuai ex- 

 periments, we see that in the first one given in the table, in which 

 the water was filtered through coarse sand taken from one of the 

 tauks of the Aquarium, whereby nearly ali the free ammonia was 

 removed, the larva) were 6.2 ^^ larger than the normal. Larva) grown 

 in water filtered through some of the same sand, which had however 

 been kept in a jar of water ten days, were increased oA% in size. 

 In the next experiment. No. 64, in which the water was filtered through 

 some of this sand which had been heated to 70" so as to kill the 

 vegetable growth in it, the larvai were slightly increased in size, in 

 spile of the orgauic ammonia present beìng nearly quadrupled. This 

 favourable effect was probably in part due to the mechanial removal 

 of suspended organic matter from the water, and in part, as we shall 

 see later, to the removal of bacteria. In the next experiment in the 

 table, No. 65, a fine sand fresh from one of the Aquarium tanks was 

 used, and the larvse grown in the water filtered through it were 

 increased hÄ%. In Exp. 99, in which a sand which had been 

 stagnating some time in a small tank was used, the larvìe were 

 only very slightly affected. In Exp. lUO an impure sand from the 

 sea shore near the harbour was used, and the ammonia in the water 



24* 



