In a field experiment on the effects of drought on spring barley the crop was protected from rain by automatic rain shelters. Various plots received irrigation at different times to give a range of drought treatments from full irrigation to no irrigation between emergence and harvest. The foliage area, light interception, stomatal resistance and leaf photosynthesis rate of five treatments were measured throughout the growing season, and a mathematical model has related the computed whole canopy photosynthesis to the measured total dry-matter yields at harvest. Hence, it was possible to estimate tha independent influences of drought on radiation interception, efficiency of use of intercepted radiation, and respiration. The analysis shows that for all treatments the decrease of intercepted radiation was the major factor in reducing yield, and it accounted for a loss of 30–40% for treatments that were stressed from the beginning of the season, and of 10–20% for treatments that were stressed after mid-May. Stomatal closure caused a reduction of up to 11% in daily photosynthesis, and the maximum effect was on plants that acquired a large leaf area before being stressed. However, the effect of stomatal closure integrated over the whole season was only 6% or less. Our measurements of internal resistance to carbon dioxide transfer were not precise enough to show significant differences between treatments; but increases of internal resistance, caused by stress, may have contributed to loss of yield.

A rotation experiment comparing the effects of length, type and management of leys on three test crops in six-course rotations during three cycles is described and the results discussed.

Although the yields of the test crops varied from cycle to cycle there were no clear-out trends to suggest that the yield improves or deteriorates as time goes on with these particular rotations; the full benefit of the leys was obtained in the first cycle.

The yield of potatoes immediately following the leys was no greater than the yield in an all-arable rotation receiving adequate nitrogen fertilizer, but the yields of two, barley crops following the potatoes were less in the all-arable rotation than in ley rotations even when 58 kg N/ha/year and 38t dung/ha/cycle had been applied and trefoil ploughed-in in the all-arable rotation. Rotations containing lucerne or lucernecocksfoot leys gave their greatest yields of potatoes and of barley following potatoes without any applied nitrogen fertilizer, but in the second barley crop showed the same sort of nitrogen responses as the all-arable rotation.

Yield data were also obtained on the leys themselves in each cycle. These showed a decline from the first to third cycle in yield of broad red clover grown as 1-year leys and in that of 3-year leys which contained lucerne, both of which could be explained by the build-up of specific diseases. There were also changes in the effect of grazing compared with mowing on the yield of some leys but little change in the effects on leys of applying nitrogen to them. However there was no suggestion that these changes affected the arable crops that followed.

Five species of rumen ciliate protozoa, isolated from sheep in which they were the only ciliate protozoon, and mixed rumen ciliates were incubated with labelled bacteria and the rate of uptake and digestion of the bacteria was followed. Evidence is presented that mixed rumen bacteria labelled by incubation with 14C-isoleucine, 14C-guanine or 35S-sulphate were engulfed more rapidly by the protozoa than those labelled with 14C-glucose but that, with Entodinium spp. at least, this apparent lower rate of uptake was associated with an increased rate of digestion. The rate of uptake of mixed rumen bacteria (at 109/ml) varied from 230 to 2670 bacteria/protozoon per hour, depending on the protozoal species, and increased with increasing bacterial population density. Starvation of the protozoa decreased the initial rate of uptake of bacteria by up to 60% and the length of time over which the protozoa would engulf bacteria. Entodinium spp. and Epidinium ecaudatum caudatum engulfed fewer bacteria when these bacteria were prepared from the same sheep as the protozoa than when they were prepared from sheep with no or other protozoal populations. Of the protozoa tested, only Entodinium spp. and the mixed large rumen protozoa digested mixed rumen bacteria, Eremoplastron bovisand Epidinium ecaudatum caudatum being inactive. It is suggested that previous estimates of the rate of uptake and digestion of bacteria based on studies in vitro were too high.

The non-rumen bacteria Eschericfoia coli and Klebsiella aerogenes and the rumen bacterium Megasphaera elsdenii were not or only poorly taken up by the protozoa. Entodinium spp. selectively engulfed bacteria of rumen origin and digested two cellulolytic bacteria rapidly. In contrast, Eremplastron bovis engulfed these cellulolytio bacteria more slowly than other bacteria. Entodinium spp. and the large rumen protozoa digested more bacterial species than the other protozoa but Butyrivibrio fibrisolvens was digested more rapidly than other bacteria by all the protozoa.

An experiment has been made using adult lactating cows in a randomized block design to study the effectiveness of basing feeding during the first 20 weeks of lactation on yield predicted from that obtained during the 8–12th days of lactation. Cows were classified as high yielding (HY) or low yielding (LY). The levels of feeding were identical within each yield group during the first 8 weeks of lactation but during the 9th to the 20th week two different levels of concentrate feed (HP and LP) were used. Over the 20-week period concentrates were offered in two or five daily feeds. Thus the effect of two levels and two frequencies of feeding on feed intake, milk yield and composition, and on live-weight change were studied.

Cows in the HY group achieved a mean daily peak yield 2·6 kg higher than predicted while those in the LY group achieved the predicted mean peak daily yield. There was no significant effect of level of feeding on the 20-week milk yield but there was a significant effect on the yield of the LYLP class in weeks 9–20. Frequency of feeding had no significant effect on yield in either high-low-yielding cows. There was no significant effect of level of feeding on milk composition but low-yielding cows gave milk with a higher fat percentage than high-yielding cows. Five-times-a-day feeding also produced milk with a higher fat percentage than twice-a-day feeding.

Live-weight changes were small and the largest variation was recorded in highyielding cows.

The number of days from calving to first service was not significantly different for the two yield groups but there was a significant effect of level of feeding.