334 Prof. E. W. MacBride. The Artificial Production of 



if possible, how far these tissues were independent of one another in their 

 modifications, or whether the modification of one caused the modification of 

 the rest. 



Description of the Experiments. 

 A. The Production of Larvce with Two Hydrocoeles. 



On thinking over the possible changes in the environment which might be 

 likely to be encountered by larvge reared in artificial cultures it occurred to 

 me that the salinity of the water in the culture-jar might be raised by 

 accidental evaporation and that this stimulus might be the factor which 

 caused the formation of a second hydrocoele. I may say in anticipation that 

 this surmise proved to be partly right and partly wrong. It is true that 

 increased salinity, acting at a certain critical period of development, will lead 

 to the production of a right hydrocoele, but other causes must also be 

 competent to produce the same result, for increased salinity will not account 

 for every instance of this phenomenon. 



It transpired, from preliminary experiments in 1914, that larvae would 

 grow fairly well in water of which the salinity had been raised by evapora- 

 tion to 37 parts per thousand, and with difficulty in water of which the 

 salinity was 39 parts per thousand. Accordingly, for all further experi- 

 ments, water of a salinity of 37 parts per thousand was employed, although 

 in 1914 some trials were made of transferring larvae which had been reared 

 up to a certain stage of development in water of a salinity of 37 parts per 

 thousand to water of a salinity of 39 parts per thousand. Water of 

 enhanced salinity is usually denoted as " hypertonic," and, for the sake of 

 brevity, I shall adopt thi§ term in further discussion of the subject. 



I was unsuccessful in fertilising the eggs in hypertonic water, although I 

 do not assert dogmatically that this difficulty could not be overcome. 

 However, I abandoned further attempts to do this, and all the eggs used in 

 these experiments were fertilised in normal sea-water, and the larvae were 

 transferred to hypertonic water when they had reached an age of three days 

 (1915, 1916) or four days (1917). The exact stage of development attained 

 at these ages varies with the time of year and with the vigour of the 

 culture, but, broadly speaking, the stage reached at four days may be taken 

 to be characterised by the completion of the alimentary canal, by the 

 division of the coelomic rudiment into two equal coelomic sacs, and by the 

 appearance of the antero-lateral and post-oral ciliated arms, but by the 

 absence of a madreporic pore. In a word, the stage may be described as one 

 in which no departure from strict hilateral symmetry has as yet occurred (fig. 1). 



In 1914, when the larvae were about a fortnight old, and had lived for 



