8 Charles Zeleny. 



The experiments were performed at the Naples Zoological 

 Station during the winter of 1902-03. The common five-armed 

 brittle-star, Ophioglypha lacertosa, was used and five series of 

 experiments, with one, two, three, four and five arms removed, 

 respectively, were kept for 46 days after the operation and no food 

 was supplied to them during the whole period. 



The resulting data show that the rate of regeneration of the arms 

 varies on the one hand with the size of the animal and on the other 

 with the number of removed arms. Medium-sized individuals 

 show the maximum rate of regeneration and there is a pronounced 

 decrease both for smaller and for larger ones. The second corre- 

 lation and the one that concerns us especially in the present paper 

 gives an increase in the rate of regeneration of an arm as we pass 

 from the cases with a smaller to those with a greater nutvber of 

 removed arms. The series with all five arms missing is excepted 

 in the statement because the animals in this lot in every instance 

 died or showed evidences of decay before the completion of the 

 experiment. 



2. Method. 



Forty-five perfect specimens were divided into five equal groups 

 of nine each, care being taken to distribute them in such a way 

 as to make the sets approximately equivalent as regards size of 

 individuals. The operations consisted in the removal of one or 

 more arms by a transverse cut at the disk level. In the first series 

 one arm was removed, in the second two contiguous arms, in the 

 third three contiguous arms, in the fourth four, and in the fifth five 

 arms. The animals were kept in ten "battery" jars, two for each 

 series and were not fed during the whole period of the experiment. 

 Measurements of the lengths of the regenerating arms were taken 

 22, ^^ and 46 days after the operation. As stated above, the 

 specimens of the series where all five arms were removed did not 

 retain their vitality for a sufficient length of time to allow of com- 

 plete comparison with the others, and they will therefore not be 

 included in the main comparison, though the data concerning them 

 are given in Table I. 



The results are given in the accompanying table (Table I), 

 which shows for each of the five series the disk diameters of the 

 specimens and the lengths of the regenerating arm or arms 22, 

 33 and 46 days after the operation. Where more than one arm 



