Íslenskar landbúnaðarrannsóknir - 01.09.1977, Side 46

44 ÍSLENZKAR LANDBÚNAÐARRANNSÓKNIR
TABLE 5. MEAN NUMBERS OF NODULES ON PLANTS IN EXPERIMENT 2.
Varieties Bacterial inoculation Lime treatments Lime High lime pellets dressing No lime Means
75 (England) 4.8 4.8 3.9 4.50
83 (Sweden) 5.5 3.6 5.1 4.73
S184 33 (Iceland) 4.9 6.4 5.0 5.43
38 (Iceland) 3.8 4.2 5.5 4.50
No inoculation 2.0 2.1 2.0 2.03
Means 4.20 4.22 4.30
No significant difference
P < 0.001
75 (England) 4.3 5.6 4.6 4.83 0.05 = ooí
83 (Sweden) 4.1 4.5 4.5 4.37 u 0.01 ~ 1
Pajbjerg 33 (Iceland) 5.8 5.1 6.4 5.77
smaíbiadet 3ö (lceiand) 4.9 5.5 5.5 5.30
No inoculation 2.1 2.6 1.5 2.06
Means 4.24 4.66 4.50
No significant difference
effective at 12°C, whereas out of the 40
Icélandic strains 12 were highly effective.
Conversely, when tested at 20 °C, 2 of the
10 Welsh strains were highly effective comp-
ared with only 5 of the 40 Icelandic strains.
The other strains were less effective at the
temperatures used.
Experiment 2.
The numbers of indigenous Rhizobium trifolii
in the area of the field experiment were
investigated using the M. P. N. method
(ViNCENT, 1970). Twenty samples were
taken from the experimental area, thoroughly
mixed together and two subsamples taken
for measuring the numbers of Rhizobium.
No nodules were found on the plants used
in this test and thus it must be assumed
that the indigenous Rhizobium were either
totally absent or in an undetectably low
number.
Number of nodules on plants in the ex-
perimental plots are summarised in Table 5.
Combined treatments were found to have
significantly different nodule numbers (P<
0,001). This difference was due to inocula-
tion with Rhizobium strains, with the unin-
oculated controls having the lowest mean
number of nodules per plant. Most of the
nodules in the uninoculated plots weresmall
and ineffective looking. The nodules in the
inoculated plots were, on the other hand,
1—2 mm in diameter, and pink because
of leghaemog;lobin. No significant difference
was found between varieties or lime treat-
ments on the formation of nodules, but it
must be remembered that the pH of the soil
was fairly high in any case (pH 5.75). No
significant interactions between individual
treatments were found.
A summarised table of shoot dry weights
for the various treatments is shown in Table
6. Surprisingly, neither inoculation nor lime
treatments had any significant effect on yield.
To test the winterhardiness of the varieties
some plants were left in the experimental
plots for the winter 1976—77. The spring
in 1977 was unusually cold and none of the
plants survived.