288 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



I73"W. 



180° 



I70«E. 



30° 



NORTH LATITUDE 



Figure 14. — Variation In trawl catch (volume in milli- 

 liters per hour of hauling) of 6-foot Isaacs-Kidd trawl 

 on five north-south sections, Hugh M. Smith cruise 27, 

 January-February 1955. 



cruise 9 in November 1951 in Hawaiian waters, 

 the average volume was 34.6 ml. and the average 

 number of organisms was 138 jDer hour of hauling. 

 Four night hauls of Smith cruise 15 in June 1952 

 at about latitude 6° N. in the Equatorial Counter- 

 current yielded an average volume of 217.9 ml. 

 and an average number of 4,803 organisms. The 

 chief difference in the composition of the catch 

 from the two areas was the much greater quantity 

 of euphausiids taken in the Countercurrent. The 

 variety and quantity of fish captured in the two 

 areas were approximately equal. Although the 

 two cruises occurred at different times of the year, 

 we do not believe that the dift'erences observed 

 were seasonal in nature, but rather were associated 

 with geographical differences in the fauna or dif- 

 ferences in standing crop of certain fauual groups. 

 Longitudinal vanations. — The 6-foot Isaacs- 

 Kidd trawl was employed on one cruise, Smith 



cruise 27, in such manner as to provide a rough 

 e\-aluation of differences in the volume of catch 

 among longitudes in the central North Pacific. 

 During January and February 1955, three to five 

 night hauls were made on north-south sections 

 along the longitudes 170° E., 180°, 173° W., 167° 

 W., and 159° W., within the latitudinal range of 

 23° N. to 37° N. As evidenced in figure 14, and 

 evaluated by an analysis of variance, the differ- 

 ences among longitudes were not significant {F= 

 1.79; d.f.=4and 16; P>0.05) ; however, the resid- 

 ual variance in the test was composed of vari- 

 ability due to latitude as well as to chance, which 

 reduced the F value for longitude. Although the 

 data are few they do show a definite trend. For 

 four of the five longitudes, the catches were gen- 

 erally low in volume between 23° N. and 30° N., 

 and moderately higher in volume between 30° N. 

 and 37° N. in the region of the transition zone. 

 Judging by the shape of the cui-ves (fig. 14), we 

 postulate that during this winter season the south- 

 ern boundary of the transition zone may have oc- 

 curred as far south as latitudes 30° N. to 32° N. 



It would have been desirable to evaluate dif- 

 ferences in the catch due to longitudinal variation 

 using a two-way analysis of variance so as to 

 separate variability due to latitude from that due 

 to experimental error. Unfortuately, unequal 

 groupings of observations along the various longi- 

 tudes made this possible for only Smith cruises 

 30 and 31. Even for these two cruises some data 

 collected at the extremities of several longitudinal 

 sections could not be used owing to lack of cor- 

 responding data on adjacent sections. Also, in 

 three instances, one of two values recorded close 

 together along a longitude section was discarded 

 in order that the analysis could be undertaken. 

 The discarded values were determined by flipping 

 a coin. 



During Smith cruise 30, in July and August 

 1955, 6 or 8 night hauls were made with the 10- 

 foot Isaacs-Kidd trawl on each of four longitudes, 

 180°, 172° W., 165° W., and 157° W., within the 

 latitudinal range of 22° N. to 50° N. (fig. 15). 

 As evaluated by a two-way analysis of variance, 

 the differences in catch among longitudes were 

 not significant (/^=3.09; d.f. = 3 and 15; P>OM\ 

 F at the 0.05 level was 3.29). As for the 6-foot 

 trawl (fig. 14), the catches taken on the northern 

 portion of each section, in this case between 35° N. 



