424 BICKFORD. [Von,. TX. 
hydranth on the oral end; both were in vigorous motion 
(Fig. 12, a, s, b). A second cut was made about half way 
down the stem, and the next morning a new hydranth (Fig. 12, c) 
was found at the oral end of the lower half. All three 
hydranths were in vigorous motion. 
The ready regeneration of this small piece seemed to indi- 
cate that the power depended but slightly upon the size of 
the fragment ; to study this point further the following experi- 
ment was made :— 
(2) Seven whole stems were placed in the first bowl, ten 
half stems in the second, twelve fourth stems in the third, and 
twenty-four eighth stems in the fourth bowl. After two days 
the following results were observed: The first stems had 
regenerated four hydranths ; the second series three hydranths ; 
the third series ten, and the fourth series twelve hydranths. 
These were allowed to grow for five days. Then the increase 
in length was measured, and found to be as follows: The 
whole stems averaged 7 mm.; the half stems 6.5 mm.; the 
fourth stems 5.5 mm., and the eighth stems 4mm. Thus the 
actual amount of growth was proportionally much greater in 
the stems when cut into eighths than in the whole stems. 
(c) While working on the pieces of stems some of the 
regenerating fragments were so small that the question pre- 
sented itself as to whether there were any special regenerating 
regions, comparable to budding zones, or to specialized regions 
of reproductive cells, as Weismann describes in his paper on 
Sexual cells of Hydro-medusae (14). Dr. Loeb had shown in 
his experiments that when stems were cut in halves the 
hydranths were formed at the same time on the oral ends of 
the fragments. An experiment was made varying from his in 
the number of sections of a stem. In this case a strong vigor- 
ous stem about 4 cm. long was cut into twenty sections ; when 
examined later, sixteen of these fragments were found with 
complete hydranths. The four small pieces which had appar- 
ently not regenerated, were studied under a low magnifying 
power, and were found to contain abnormal hydranth forms. 
In experiments previous to this, numerous cases of hetero- 
morphosis had been found, as nearly all fragments regenerated 
