Table 22-1. Numbers of Larvae Obtained by 
Sampling Different Depths Across an Estuary 
Stations on 
Cross Section 
Number of Larvae 
Surface 
Mid-depth 
Bottom 
la 
_ 
— 
— 
lb 
— 
— 
— 
1 c 
— 
— 
— 
2a 
— 
— 
— 
2b 
— 
— 
— 
2c 
— 
— 
— 
3a 
— 
— 
— 
3b 
— 
— 
— 
3c 
— 
— 
— 
4a 
— 
— 
— 
4b 
— 
— 
— 
4c 
— 
— 
— 
5a 
— 
— 
— 
5b 
— 
— 
— 
5c 
— 
— 
— 
The data in Table 22-1 illustrate an outline for triplicate determinations of a 
given ichthyoplankton larval species for five locations along the cross-section of 
an estuary at three depth levels. Since each station location has three replicate 
samples at each depth, the classifications “depths” and “locations” are 
completely crossed, while the triplicate determinations provide three 
replications for each combination of depth and location. In the above table 
there are three columns corresponding to depths, five rows corresponding to 
the five locations, and there are 15 cells each containing three replicate 
observations. 
In the general case there are xy a observations, i = l,...,p rows, j = l,...,q 
columns, a = 1,...,?? replicates. The model for the analysis is assumed to be: 
x ij a = M + I + ?)j + Xy + Sjj a • 
In the above model, ju is the mean, £ represents row effects (location 
variability), and r?j represents column effects (depth variability). The 
interaction effects Xy represent any variations which may be peculiar to a 
particular combination of station and depth, and the effects 5^ are normally 
distributed random components with average value zero for each ij. 
340 
