Two examples are given of field trials for which randomized blocks would seem the most suitable design, but in which special considerations of the future of the trial or the nature of the treatments make complete randomization impossible or impracticable. A type of design which takes account of these difficulties is described, together with the appropriate methods of analysis.

The author wishes to thank Mr S. C. Pearce for much valuable advice throughout this investigation.

1. A series of experiments was carried out to assess the importance of the sense of smell to the grazing sheep.

2. The results of the first two experiments described indicated that the sense of smell is of importance in the initial stimulation of appetite, but under the experimental conditions the selection of particular herbage species was not affected.

3. Further experimentation showed that, because of its adaptation character, olfaction can only be of supplementary importance in influencing the food selection of the grazing sheep.

Since sheep failed to associate either a red or blue colour with the palatability of their food during the course of an experiment designed to test for colour vision, it was concluded that sheep are colour blind.

1. Data are presented on the volume and other dimensional relations of structures in the skin of four sheep of the Romney Marsh breed, namely a lamb aged 1 month, a ram aged 2½ years and two 8-year-old ewes. All the animals were in good condition when the skin samples were obtained, yet the skin of both ewes contained structures believed to be degenerating follicles.

2. A method was developed by which it was possible to measure the volume of follicles and other structures in the skin, with an error of less than 1%. This method involved the tracing of projected images of horizontal sections of a selected follicle bundle throughout its depth in the skin, and measurement, by means of a planimeter, of the areas of structures shown on each tracing. The volume was calculated from the sum ofall the areas multiplied by the thickness of the sections. A method was also evolved of measuring the curled length, as distinct from the ‘depth’, of a follicle. The difference between the length and depth of the majority of the follicles in these samples was less than 10%.

1. Methods for the determination of lignin which involve the use of 72% sulphuric acid have been reviewed.

2. Three methods in common use, viz. those of Norman & Jenkins (modified Common), Crampton & Maynard (modified Lancaster) and of Ellis, Matrone & Maynard were studied, and comparisons made of their relative effectiveness.

3. That of Crampton & Maynard was considered less satisfactory than the modified Norman & Jenkins procedure, on the grounds that it contains several sources of error, the most serious of which are the use of formaldehyde and the failure to remove pentosans.

4. The yields of lignin obtained with the method of Ellis, Matrone & Maynard are appreciably higher than those resulting from the modified method of Norman & Jenkins, by reason of the failure to use a protein correction factor. When such a correction is made, results from the two methods are comparable.

1. It has been shown in a previous paper (Callow, 1948) that, with ‘standard’ carcasses of cattle, sheep and pigs containing more than 18 % of fatty tissue, FT/C, there is a close correlation between FT/C and MT/C, the percentage of muscular tissue. This is shown by the linear regression equation

MT/C = 76·1 — 0·684FT/C ± 1·0 (r= — 0·9782).

The reasons for this close correlation and the effects of varying rates of fattening on the constants of the generalized equation

MT/C = a — bFT/C

have been deduced by a mathematical and biological analysis of the data.

Fertilizer placed in contact with the seed damaged the germination of sugar beet, mangolds, swedes and peas in 1947; damage was also caused by fertilizer placed directly below the seed. In 1947 and 1948 some damage to the germination of sugar beet was caused by fertilizer in bands 3 in. below the soil surface and 1 in. to the side of the seed; in both years bands of fertilizer 3 in. to the side of the seed were safe. Where placement of medium or full fertilizer dressings is adopted for row crops, the bands should be placed not less than 2 in. from the seed to avoid the risk of injury to germination.

All methods of placing gave consistently higher yields of peas than broadcasting fertilizer, the average advantage of placing being over 2 cwt. of peas per acre. In all the experiments on peas in 1948 early applications broadcast and worked deeply into the seed-bed were somewhat superior to late applications made at sowing time.

1. The object of this investigation has been to advance knowledge of the methods necessary for the practical application of the transplantation of fertilized ova in cattle.

2. The induction of ovulation at any specific time can be produced by squeezing out the corpus luteum per rectum. Oestrus occurs from 2 to 4 days afterwards; this is accompanied by ovulation and the ova so produced are readily fertilized. The onset of oestrus after the expression of the corpus luteum is hastened by previous treatment with P.M.S.

3. The induction of multiple ovulations is necessary to produce an augmented supply of fertilized ova, but it is of primary importance that the ova so produced are fertilized, viable and capable of full development. Four methods of doing this have been tested.

The Journal of Agricultural Science

Vol. 38, Part 3

July 1948

Page 323

Fructose Content and Fructolysis in Semen. Practical Application in The Evaluation of Semen Quality

By T. Mann, Molteno Institute, University of Cambridge

The last sentence on p. 323 at present reads as follows: “0.25M phosphate buffer pH 7.4, prepared by dissolving Na2HPO4 (or Na2HPO4. 12H2O) in 190 ml. glass distilled water and adding 10 ml. IN-HCI.”