The Relations betweea Marine Animai and Vegetable Life. 343 



was able to maintaiu an aquarium with the sea-water uuchanged 

 for months K 



An inquiry into the conditions and inter-relations of marine animai 

 and vegetable life bas thus a veiy practical bearing, but, quite apart 

 from this, it is of interest to know somethiug of the processes taking 

 place on a vast scale in the open sea. What are the most eifective 

 agencies in the removal of this nitrogenous waste material excreted 

 by the animals, and under what conditions do they act? Are they 

 chiefly or almost wholly of a vegetable nature, or are other important 

 factors at work? Such are the questions it is bere attempted to 

 answer. 



Method of Experiment. 



In the present inquiry it was found necessary to attack the 

 problem from a triple standpoint, from the chemical, the physiological, 

 and the bacteriological. The physiological part of the research 

 consisted in the continuation of the method described in the above 

 mentioned paper. In this method, the larva; of the sea-urchin 

 Strong^jlocentrotus lividun, which are readily obtained from artificial 

 fertilisations, are grown in the various specimens of water, obtained 

 under various conditions, and then after eight days' growth are killed 

 and preserved, and measured in series of fifty under the microscope. 

 From the mean size of the larvai obtained under these various con- 

 ditions, as compared with that of the larvai obtained on growth under 

 normal conditions in ordiuary tank water, it was possible to deduce 

 conclusions as to the purity or otherwise of the water under examina- 

 tion. In every case two measurements of each larva were made, 

 namely of the body length, and of the aboral arm length. These 

 latter were then always calculated as percentages on the former. 

 The mean values obtained for these measurements are given in the 

 table at the end of the paper. They are given in scale units of the 

 micrometer eyepiece. If it be desired to reduce them to millimetres, 

 they must be divided by 152.3. In the table it will be seen that 

 altogether 140 series of measurements, each comprising fifty larvse, 

 were made. The few experiments marked with an asterisk were 

 made upon the larvse of Sp//fsrec//inus or Eclimus^ and not those of 

 Strongìjlocentrotus. In addition to measuring the size of the larvse, 

 the numbers of ova used in each case, and the number of blastulse 



* The Aquarium 1854 pag. 13. 



23* 



