August 29, 1902.] 



SCIENCE. 



345 



variations, chiefly in radial canals and go- 

 nads. Ehegmatodes tenuis, a leptomedusa, 

 presents a much higher ratio of varia- 

 tion, no less than twenty per cent, show- 

 ing more or less marked variation in char- 

 acter and size of radial canals. In this 

 species a rather prominent variation was 

 that of the bifurcation, looping and an- 

 astomosing of the canals, suggesting 

 a probable means by which the large 

 nimiber of canals may have arisen. 

 The speaker has already called atten- 

 tion to a similar condition in Gonion- 

 ema as has Mayer in Pseudoclytia pen- 

 tata. Variation of the gonads followed 

 a similar range, but in view of the com- 

 paratively immature condition of many of 

 the specimens, no attempt was made to go 

 into details upon this point. 



The Sense of Taste in Fishes: C. Judson 



Heerick, Denison University. 



Sense organs known as terminal buds, 

 which resemble taste buds in structure and 

 are wholly independent of the lateral line 

 system of sense organs, are found freely 

 scattered over the entire body surface of 

 the catfishes. Like the taste biids in the 

 mouth, they are innervated by communis 

 nerves and presumably they serve the gus- 

 tatory function. To test this hypothesis 

 experiments were performed to determine 

 how the siluroid fishes perceive their food. 

 It appears that the sense of sight plays very 

 little part, the sense of smell and the sense 

 of touch considerable parts, but the sense 

 of taste clearly the chief part in their de- 

 tection of food. These fishes appear to 

 taste not only in the mouth, but by contact 

 with sapid substances by the barblets or 

 the skin of the body at any point as far 

 back as the root of the tail fin. Gustatorj^ 

 and tactile sensations arising in these cu- 

 taneous areas commonly cooperate in evo- 

 king the reflex of seizing food, but by train- 

 ing the fishes can be taught to discriminate 



between the gustatory and the tactile ele- 

 ments in the stimulus and to respond only 

 when both are present, ignoring simple tac- 

 tile contacts. Some other fishes with large 

 eyes and having these terminal buds less 

 generally distributed over the body surface 

 feed in a wholly different way, by snap- 

 ping up a moving bait (visual stimulus), as 

 every fisherman knows. (Discussed by 

 Messsrs. Dyche, Eigenmann and Surface.) 



On the Value of an Apparently Fixed 

 Food-Hahit in Scale Insects as Determin- 

 ing Species: Charles Lester Maelatt, 

 U. S. Dept. of Agriculture. 

 The speaker detailed striking variability 

 in the food habits of certain species of scale 

 insects ; food plants of the same species in 

 different districts may be multiple or sin- 

 gle, and may change from year to year or 

 after a series of years, the plants once at- 

 tacked becoming practically exempt and 

 new host plants being assumed. A fixed 

 food habit could not, in the opinion of the 

 speaker, be considered as necessarily of 

 value in determining species. (Discussed 

 by Messrs. Webster and Hopkins.) 



Statistical Study of Variation in the Peri- 

 odical Cicada: Herbert Osborn, Ohio 

 State University. 



A distinct variation from the typical 

 form of Cicada {Tihicen) septendecim has 

 been recognized since 1829, and was de- 

 scribed in 1851 as a distinct species, cassini. 

 Riley, however, and later Marlatt have con- 

 cluded it should be ranked simply as a di- 

 morphic variety. In the occurrence of the 

 present year this form has been so very 

 abundant that a statistical comparison of 

 the two forms was naturally suggested. 

 Measurements of 800 specimens show a very 

 decided constancy of each variety and for 

 each sex of each variety. This constancy 

 can best be shown in curves of frequency. 

 Color variation is also very constant and is 

 believed to be as constant as the other fea- 



