For each test the strength of the response according to the observer's 

 visual impression and also the strength of the response as measured quantitatively 

 is given in table 1. The variate for the latter, which was used for statistical 

 analysis, may bo explained by means of an example. In the experiment conducted 

 with an extract of skipjack flesh on Thursday of the second week, when the fish 

 had been fed skipjack flesh (see table 1, E2D3F2) » the mean time spent by the fish 

 in the attraction area was 7.2 seconds during control periods and 28.6 seconds 

 during test periods, with a mean difference of 21.4 seconds. The mean number of 

 "passes" in the attraction area was 4.4 during control periods and 15.2 during 

 test periods, with a mean difference of 10.8 passes. The differences in both time 

 and count reflect the response, althovigh their relative value as a measure of res- 

 ponse is difficult to assess. The two were added, giving 32.2; rounded off, 

 giving 32; and Increased by 20, giving 52. The last step, adding 20, was included 

 because in some cases negative differences were obtained, i.e., the fish spent 

 relatively more time and/or made more passes in the attraction area during control 

 than during test periods. 



Order of testing la not indicated in table 1. Inspection of the data 

 indicated no consistent trends in response with order of testing within days: on 

 some days the first substance tested gave a high response whereas on other days 

 the last substance tested gave a high response. Similarly, Inspection of the data 

 indicated no significant differences between the preserved and non-preserved blend. 

 Accordingly the results were analyzed in toto according to three criteria of 

 classification: 



(1) Extract — E]^-squld; i;;2-9kipjack; Eg-shrlmp; E4-blend 



(2) Days — D;i^-Tuesday ; Dg-Wednesday; Ds-Thursday 



(3) Food (= Weeks) — Pi-squid; F2-sklpjaok; F3-3hrimp 



Table 1. — The response of tunny to four extracts (E) on 

 three days (D) in each of three weeks during 

 which they were fed three different foods (F) . 

 For further explanation see text. 



