408 
R. P. COWLES 
with the latter. If a piece of coral was placed under the horizontal 
screen the behavior of the opbiuroid was almost invariably changed. 
Instead o^ moving through the shaded region, the specimen 
put one or more of its rays against the coral, tilted its disc and 
remained quiet as though the conditions for its comfort were 
satisfied. 
EFFECT OF LIGHT AND CONTACT STIMULI ON DIRECTION OF 
RIGHTING 
The ophiuroids with which I have worked do not seem to show 
any permanent habit of using a certain pair of rays on which to 
right themselves. That is, when one of the creatures with rays 
of about equal length is inverted in a dish of water, care being 
taken to rule out light, handling, etc., no one pair seems to have 
greater functional value than another. Normal structural charac- 
ters do not seem to determine the rays used, but when such exter- 
nal factors as light, contact with solid walls and handling are 
allowed to operate, they seem to be of much importance in deter- 
mining the direction of righting and consequently the rays used. 
Undoubtedly internal physiological conditions do at times influ- 
ence the behavior, but there does not seem to be a permanent 
habit of righting on a certain pair of rays persisting from day to 
day. 
In the series of experiments from which the following typical 
cases are taken, the same glass dish filled with sea-water was 
placed in front of a wide window as in the previous experiment, 
so that the strongest light came from the side indicated by the 
straight arrow in the figures. An ophiuroid was placed in the 
water and allowed to come to rest with one of its rays in the corner 
a. The rays were then numbered, as shown in fig. 10, ray 1 being 
the ray in contact with the corner. When the ophiuroid was taken 
from the position a, inverted and placed at the position b, so that 
therayl was pointed in a diametrically opposite direction, the crea- 
ture righted itself on rays 3 and 4, returning directly to its original 
position a. Similar trials were made many times, and as a general 
rule the behavior was the same as that just described. Now the 
