6 University of California Publications in Zoology [Vol. 19 



The animals were kept in a large glass dish in the laboratory ; and 

 as a rule individuals were picked out of this stock as needed, and 

 thereafter each one was kept in a finger bowl. Some species could 

 be kept alive for a week, but most of the forms did not withstand 

 laboratory conditions well. While I thought at the beginning that it 

 would be desirable to test each one of a given set of individuals for 

 several days, if possible, that plan was soon abandoned, and many, 

 if not most of the experiments were performed with animals col- 

 lected during the morning of the day on which they were used, or 

 during the preceding night. But if specimens appeared to be in good 

 condition on the second day I did not hesitate to use them. The 

 precautions taken to have fresh material are not necessary, in my 

 experience, to guard against alterations in the form of the responses. 

 I believe that there is something connected with laboratory conditions 

 that does affect the reactions of some species so that there is a change 

 after the animals are removed from the ocean. But if the change 

 occurs it appears shortly or within a few hours after the animals are 

 brought in. Results are uniform after that, so far as sign of reaction 

 is concerned, as long as there is any response. 



The water in which the animals were kept after collection was 

 either that from the laboratory system or it was brought into the 

 laboratory from the end of the pier (1000 feet from shore) in glass 

 jars. So far as I could determine from a large number of "trials with 

 different species, the behavior is practically the same in water from 

 the two sources; though I believe that the animals remain in better 

 condition and live longer if they are not put into water that has 

 passed through the pipes of the pumping system. 



The experiments were carried on with adults of the following 

 forms: the copepods Acartia tonsa and A. clausi, Calanus finmarchi- 

 cus, Eucalanus elongatus, Labidocera trispinosum, Metridia lucens, 

 and the chaetognath Sagitta bipunctata. It was originally intended 

 to carry out the same series of tests with specimens of each species, but 

 this proved impossible. The chief reason is that some species were 

 not obtained in sufficient numbers or long enough through the year. 

 Acartia tonsa, Calanus, and Sagitta could be counted on at almost any 

 time, and were obtained in fair abundance, but the other species were 

 more or less erratic in occurrence. There were times, indeed, when 

 collecting was uniformly unsuccessful, even Acartia being absent from 

 the surface at night though ordinarily very abundant then. The col- 

 lections varied from time to time, also, so that it was necessary to use 

 different species on different days as they happened to come. 



