168 W. J. CROZIER 
either case the same; when slowly swimming the de-eyed hamlet 
will most often neatly avoid contact with a rod or wire situated in 
its path, but more successfully if the end or edge of the rod pre- 
sents a sharp corner. Similarly, in many cases, the fish is some- 
what better stimulated by a thin wire (less than 1 mm. in diame- 
ter) than by a thicker one and by a rod of square cross section 
than by one of similar size (several centimeters in diameter) 
but with a smoothly rounded end and circular cross section. 
The inference from these tests is, unavoidably, that mechanical 
deformations in the water, of a somewhat zrregular character, are 
the means of stimulation. It was shown by appropriate elimi- 
nation experiments that the nostrils and lateral-line organs could 
not be concerned, and this is further made obvious from a con- 
sideration of the local nature and manner of distribution of these 
reactions over the body of the fish.’ 
The mode of excitation in these reactions is in certain par- 
ticulars significantly different from that in some reactions which 
have previously been attributed to tactile excitation of the skin 
in teleosts (cf. Parker, ’04, pp. 61, 62; Jordan, 17). A current 
from a pipette or ripples at the water surface frequently failed 
to induce any perceptible reaction in a de-eyed hamlet, al- 
though immediately after this, or immediately before, a slender 
rod or wire slowly brought to within 5 em. of the snout or caudal 
peduncle led to well-defined reactions. Moreover, it was often 
possible to get good reactions to a thin rod in water much dis- 
turbed by a current of relatively large volume. 
The snout and lips of the hamlet were the most sensitive re- 
gions of the animal’s surface. There is thus a general parallelism 
between the distribution of this delicate tactile sensitivity and 
that of skin sensitivity to currents, as described by Jordan (’17). 
Whether or not this indicates the actvity of the tactile corpuscles 
in the reactions herein discussed, I am not sure; but I suspect 
that the tactile corpuscles may not be involved, although con- 
It may be suggested that the reactions of Amoebae to insoluble substances, 
as described by Schaeffer (’16), are possibly due to some such form of irritability 
as that herein considered. Certain peculiar phenomena obtainable with human 
erythrocytes (Oliver, ’14; Kite) are also suggestive in this connection. 
