COLE. — IMAGE-FORMING POWERS OF VARIOUS TYPES OF EYES. 351 



been kept in the dark, were tested, each in its turn, first, with the lights 

 separately, and then with both acting at the same time. In each ex- 

 periment the worm was placed in the normal position, — that is, with 

 its long axis at right angles to the rays of light (see p. 346), — and al- 

 lowed to crawl a distance of 15 cm., i. e., until its anterior end crossed 

 the outer circle marked on the table (see p. 346, and Figure 2, p. 34.S). 

 The course and the angle at which the worm crossed the circle were 

 immediately recorded on blank diagrams differing from the table dia- 

 grams only in the absence of one of the three circles. These circles 

 were divided, as on the table diagram, into arcs of 10° each. When 

 the worm had been tested five times with the head started in one direc- 

 tion, it was then tested an equal number of times with the head started 

 in the opposite direction. Experiments of this kind and number were 

 made in succession with 

 five different worms, such 

 experiments — 50 in num- 

 ber — constituting a set. 

 Three such sets of experi- 

 ments were made, and they 

 may be designated as A , B, 

 and C respectively. Sets 

 B and G were performed 

 under precisely similar con- 

 ditions, and may be consid- 

 ered together ; in set A 

 the conditions differed from 

 those of sets B and C only 

 in the method of starting 

 the worm. This method 



proved to be much less satisfactory than the one employed in sets 

 B and C. 



In the experiments of set A, a ground-glass plate (Figure 3, P) was 

 placed on the diagram on the table, and the worm was allowed to crawl 

 on the roughened glass surface. The plate was kept moist by the 

 frequent use of a damp sponge, and when thus moistened was suflft- 

 ciently transparent to show the circles and radii on the table below. 

 Two strips of thick glass (.s, s') — made opaque by being painted, except 

 "where the worm was to come in contact with them — were placed par- 

 allel to the normal axis, with one of their ends at the directive axis, 

 and with a space between them only a little wider than the diameter 

 of the worm's body. This formed a trough, or runway, in which the 

 worm was placed and allowed to start crawling, whereupon a third 



■M^;^-^}'^^^mS^W^^M 



Figure 3. Apparatus used in earlier experi- 

 ments witli earthirorin : P, ground-glass plate; 

 s, s', small glass strips forming runway. 



