46 
PACIFIC SCIENCE, Vol. XX, January 1966 
FlO. ,3. Relationship between logarithm chlorophyll a (C) and logarithm zooplankton (Z) for the 36 sta- 
tion-pairs of Table 1, A and B (AB-36 series). Antilogarithms are in mg/m 2 over 0-100 m and ml/10 3 m 3 
over 0-300 m. Station-pairs in series AB-11 and AB-8 are identified as in Fig. 2. The solid line (— — -) is 
the fitted structural regression of Z on C and the dashed lines (- — ) are the 95% confidence limits of the 
regression, both for the AB-36 series; the dotted lines ( ) are the 95% confidence limits of a similar 
regression for the AB-11 series. 
falling within the 95% confidence region for 
either of the regressions and just outside this 
region for the other regression. There were 8 of 
these, making 1 1 station-pairs for which the ob- 
served Z and M were both related to C in a way 
close to the real relationships. The set of data 
from these pairs is henceforth called AB-ll-Z. 
The localities are shown by circles (with or 
without enclosed dots) in Figures 2, 3, and 4. 
Correlation coefficients for these 11 station- 
pairs show that Z and M are related to C, Z 
independently of M, but M not independently 
of Z; and related to each other through their 
common association with C, but not otherwise 
(Table 3, "selected data”). The regressions on C 
for the 1 1 pairs, corresponding to ( 1 ) and ( 2 ) 
for 36 pairs, are 
Z ~ 1.9707 + 0.619 (C -1.4287) = 1.0863 
+ 0.619 C... (3) 
M = 0.8665 + 0.666 (C -1.4287) = -0.0850 
+ 0.666 C. . . (4) 
with 95% confidence limits of the regression 
coefficients respectively 0.658 ± 0.152 (0.506 
to 0.810) and 0.773 ± 0.522 (0.251 to 1.295). 
The 95% confidence limits for the regression 
relationships are shown as dotted lines in Figures 
3 and 4. 
The similarity of the two regression coeffi- 
cients for AB-36-Z, and likewise for AB-ll-Z, 
suggest a coefficient near 1.0 for the regression 
of M on Z. Such values (1.054 for AB-36-Z, 
1.077 for AB-ll-Z) are obtained when points 
are grouped for Bartlett’s method in the same 
