horizontally would take them away from roots. Therefore, CO 2 seemed the
prime candidate as a root signal.
Subsequent investigations of responses to salts, usually unbalanced
salts, in or on gels, sand and soil (Prot, 1978a,c, 1979a,b; Bilgrami
and Jairajpuri, 1984; Riddle and Bird, 1985; Castroet al., 1990, 1991;
Abou-Setta and Duncan, 1998), made salt gradients appear increasingly
important. Experiments on long-distance movement by many species
(Rode, 1962; Prot, 1980; Thomas, 1981), including the notoriously
sluggish juveniles of root-knot nematodes (Johnson and McKeen, 1973;
Prot, 1975, 1976, 1978b; Prot and Netscher, 1978; Prot and VanGundy,
1981; Dickson and Hewlett, 1986; Pinkertonet al., 1987), showed that
nematodes frequently moved more than 15 cm and sometimes 1 m in
less than 1 month. Movement by the root-knot nematodeMeloidogyne
javanica towards moisture in tubes of sand was reversed when the
concentration of Hoagland’s plant nutrient solution and other salts was
highest at the wet end (Prot, 1979b). Salts affected different species some-
what differently, however, and, while two root-knot nematodes were
repelled by a wide range of salts,Heterodera oryzaewas repelled only by
sodium, andScutellonema cavenessiwas unaffected. Movement away
from salts vertically would usually lead nematodes to deeper regions of
higher water content. The role of salts as root-finding cues, however, was
not clear, because sodium, which occurs in highest concentrations near
roots, was repellent.
During the late 1980s and 1990s, temperature and CO 2 gradients were
re-examined under more stringent conditions than in the 1960s. The
extreme sensitivity ofM. incognita(Dusenbery, 1988b; Plineet al., 1988)
to small temperature changes (0.001°C) rekindled the hypothesis that
metabolic heat from roots could attract nematodes (El-Sherif and Mai,
1969). Laboratory simulations of the heat waves that move through
natural soil every day as a result of surface heating and cooling confirmed
that heat waves can greatly alter the vertical distribution of plant
nematodes within the root zone within hours, and at least two species
responded by moving in opposite directions (Robinson, 1994), consistent
with computer models of nematode movement (Dusenbery, 1988a,c).
Nematodes were attracted to minute sources of CO 2 in the soil and
behaviourally relevant release rates of the gas from a point source were
shown to be achievable by roots and other biological sources (Robinson,
1995). In pot studies, ammonium was shown to repel nematodes from
tomato plants (Castroet al., 1991).
Perry (1997) suggested that signals affecting nematodes in the soil are
likely to be water-soluble, facilitating the establishment of concentration
gradients in soil water. They are also likely to be volatile, and many
substances are both. As noted by Campbell (1985), the fact that
diffusivities in the liquid phase are four orders of magnitude smaller than
those in the gas phase indicates that respiratory gas exchange in a soil
profile without a continuous air phase is, for all practical purposes, zero,98 A.F. Robinson