untitled

(Brent) #1
Hypothetically, additional salt in the diet of quail might affect their physiology and
hence their dynamics, particularly in sodium depleted areas, and the same may be
true of supplementary feeding with wheat. However, suppose that supplementary feed-
ing has an effect only when there is adequate salt in the diet. In such circumstances
two separate experiments would produce the fallacious conclusion that, whereas salt
has an effect, wheat has none. The interactive relationship between the two factors
would have been missed and the resultant management would have been inappro-
priate. One looks for an interaction by calculating whether the effect of the two fac-
tors in combination is greater or less than the addition of the two effects when the
factors are evaluated separately. That is achieved by ensuring that each level of the
first factor is run in combination with each level of the second. The factors are said
to be mutually orthogonal(at right angles to each other).
The design logic (Fig. 16.3) is seen as a simple extension of the logic of the one-
factor design and the design layout continues to adhere to replication and interspersion
of treatments. Since there are now eight treatments, each with two replicates, the
interspersion of treatments can best be achieved by laying them out either in a
systematic manner as with a Latin square or, as in the example, assigning their posi-
tions on the ground by random numbers.

EXPERIMENTAL MANAGEMENT 277

Factor: WHEAT (4 levels)
SALT (2 levels)
Response variable: Density or rate of increase of quail

Design logic

Rep
Rep

Rep
Rep

Design layout

Rep
Rep

Rep
Rep

0 WHEAT 30 WHEAT 70 WHEAT 250 WHEAT (kg / ha)

Rep
Rep

Rep
Rep

Rep
Rep

Rep
Rep

No SALT

SALT

250
WHEAT
No SALT

0
WHEAT
No SALT

30
WHEAT
No SALT

0
WHEAT
No SALT

0
WHEAT
SALT

70
WHEAT
SALT

30
WHEAT
No SALT

250
WHEAT
No SALT

0
WHEAT
No SALT

70
WHEAT
No SALT

250
WHEAT
SALT

70
WHEAT
SALT

30
WHEAT
SALT

250
WHEAT
SALT

70
WHEAT
No SALT

30
WHEAT
SALT

Fig. 16.3Two-factor
experimental design.
Do the two factors act
independently of each
other or do they
interact? If the latter,
do they reinforce each
other or oppose each
other?

WECC16 18/08/2005 14:47 Page 277

Free download pdf