The figures refer to maize growing, and incidentally to the cost of ox labour. The latter point may conveniently be taken first.

Oxen can be used for some years, and then fattened up for the market and sold, often at a higher price than they cost in the first place. This is apt to make the farmer think that ox labour costs nothing, or next to nothing. A more detailed examination does not confirm the view, however. There are, of course, profits to be made from breeding stock, but to answer the immediate question we must consider trek oxen apart from other cattle.

It is true that depreciation is trifling. If, for example, a four-year-old animal costs £12, and can be used for, say, five years, then fattened up and sold at £15, we need to know the losses to allow for meanwhile. A good farmer of my acquaintance puts the loss at one animal in eight; if, however, one allows for the inferior conditions prevailing on many farms, and the possibility even though remote of cattle plagues, the loss is probably higher. If it be one in five, the value of the four remaining cattle just equals that of the five when bought, and the depreciation is nil.

An ox will require for pasture perhaps two morgen of good ground, worth £8 per morgen, to four morgen of poor land, worth £4 per morgen, or £16 worth of ground in either case, and even then will probably cost something for winter feed besides. There is also capital expenditure for dip, water supply and fencing, which we will put at £160 for a herd of forty. Thus we have:

32 at 7 per cent. interest = £2.24

Depreciation on Dip, etc., say, 5 per cent. on £4...

Further, the annual expenses:

One Herd Boy (£30 a year) for 40 Cattle
Winter Feed, grown or purchased, say
Dipping and other Expenses

or £3 10s. per ox per year.

...

=

0.20

=

0.75

0.25

0.05

£3.49

It requires pretty good management to get 140 days' actual work per beast, especially remembering the proportion of older ones that have to be fattened up for the market. Taking that figure, the cost comes to about sixpence per working day.

The foregoing figures are not taken precisely from any one farm, but are rather of the nature of guesses based upon experience; in the second example-on maize growing-the figures are the actual averages for four farms in the Orange Free State and Transvaal for the season 1916-17.

Current Expenses:

Seed (partly purchased, but including also the value of
maize grown on the farm and reserved for seed)
Fertiliser

Native Labour (including rations, medical attendance, etc.)
Animals (cost of fodder, dipping, etc., but excluding
capital costs-see below)

Contract Work

(Two of the farms are managed by the owners, two by paid managers. In the latter case the actual salary and commission is taken. On one farm practically nothing but maize is grown; on the other the manager's time is about equally divided between maize-growing and stock-breeding, so half the salary is debited to maize. On the other two farms an amount corresponding, as closely as possible, in accordance with the amount of work done has been taken. If anything, the figure stated is an under-estimate, as it would be an exceptionally good manager who would keep either of the owner-farms up to its present standard of cultivation.)

£255

789

1,910

232

117

835

159

1,093

£5,390

(The yield was 17,119 bags, giving a total cost of 9s. per bag. The excess of the selling price above this is profit in the strict sense. The average selling price was about 12s. 6d.)

It should be remarked, however, that this estimate of costs is derived from some of the most skilfully managed farms in the country, so that it is improbable that the cost of production is so low on most farms. If these farms had been in the hands of the most penurious owners, the saving in managerial expenses could hardly have amounted to £700, or one-eleventh of the total cost. This advantage would have been lost if the yield, instead of being 19 bags per morgen, had fallen to 17 bags. It is hardly necessary to say that inferior management would certainly have resulted in a lower yield than this, so that good and well-paid management is clearly economical.

CALIBRATION OF GERBER MILK BUTYROMETERS.

By C. O. WILLIAMS, B.Sc., A.R.C.S.,

Lecturer in Chemistry, School of Agriculture, Cedara, Natal.

Read July 17, 1920.

(Abstract.)

The object of the investigation was to ascertain if the centrifugal method of calibrating butyrometers, given by Day and Grimes in vol. xlviii of the "Analyst, is sufficiently accurate when making use of ordinary commercial paraffin oil and centrifuging the butyrometers to the same degree as is done when testing milk samples.

The butyrometers taken for this investigation were first calibrated by the standard gravimetric method, using mercury, and the error per unit graduation ascertained in each case. The calibration was repeated by the centrifugal or paraffin method, and the unit error in each case obtained, as in the previous method.

It was noticed that the difference between the unit errors obtained by these two methods was approximately a constant amount (0.017 per cent.) for each butyrometer calibrated. This constant difference is apparently due to the incomplete separation of the water and paraffin columns during the process of centrifuging. Therefore, when adopting the centrifugal method of calibration, it is necessary to multiply this constant by the observed volume of the paraffin column and subtract this product from the observed total error.

On testing a further lot of butyrometers by both methods and applying the correction in the case of the paraffin method, results were obtained that were very concordant with those obtained by the standard mercury method.

Lastly, it is pointed out that the correction obtained in this investigation would probably only apply when using the same brand of paraffin and centrifuging to the same degree, so that each investigator should work out a similar method and calculate the correction for himself.

NOTE ON ROCK-GRAVINGS AT METSANG,
BECHUANALAND PROTECTORATE.

By A. J. C. MOLYNEUX, F.G.S.

Read July 15, 1920.

In the report of the South African Museum for 1918 and 1919 the Director, Dr. Peringuey, refers to the existence of some gravings on rocks in the South-West Protectorate, not representing as usual animals or figures, but hoofs of animals and human feet: gravings of a new type. Casts had been obtained for the Museum, of which photographs are given in the report. Dr. Peringuey repeats a statement by the late Theophilus Halm "that in the Namib the Bush People had signs (?) painted (?) or graved (?) to denote places where they had found water.'

I have a note of seeing, in August, 1913, somewhat similar gravings at Metsang, in the Bechuanaland Protectorate, some three miles west of Pilane Siding (mile-post 984), eight miles south of Mochudi station. At this siding an escarpment of Waterberg sandstone, facing the south, crosses the railway and merges into higher land to the west. At Metsang is a clean, level rock floor of grit, some 50 yards in diameter, in the centre of which is a hole or enlarged joint six feet across and about the same measurement in depth, but would probably be deeper were it cleaned out. It was dry at the time of my visit, but that it had been greatly frequented was shown by the polished condition of the rocky rim, brought about by the passing of feet. All around the bare floor the country is of loose red sand, so that accumulations of water, even in the wet season, are uncommon.

On the flat surface are chipped or graved the outlines of human feet of sizes ranging from those of children to that of a large adult, which measured 14 inches long. There are also the drawings of pads of canines, lions and smaller felines, to the number of about thirty. Some are faint and others sharply cut-suggesting that the gravings were done at different times. There are none of antelopes or other animals. In the originals there is no attempt to show relief-and in this they are different from those shown by Dr. Peringuey. Native legend is that the gravings were done by "ancient people."

NOTE. Since writing this note I have been reminded of a paper appearing in the " Proceedings of the Rhodesia Scientific Association," 1907, vol. vii, pp. 59-61, entitled "Ruins at Bumbusi " (Wankie District), in which illustrations of rock gravings of antelopes' hoofs are given.

BA-VENDA.

By REV. H. A. JUNOD.

Read July 17, 1920.

I had the opportunity of spending a few weeks amongst the Ba-Venda during the last summer. For a long time I had wished to visit the Zoutpansberg mountains. I had heard of marvellous ferns growing in splendid gorges, and I wanted to pluck them and dry them for my herbarium. Full of expectation, dreaming of botanical specimens new to science, I reached the Gooldville Mission Station at the end of January, with a lot of sheets of desiccating paper. But the rain fell; it fell and fell again. The rivers were soon filled, and filled to such a height that it was impossible to travel, impossible to reach those picturesque hills which stood calmly in the background of the country in a provoking attitude, most of the time hiding themselves in impenetrable clouds. I had to renounce definitely the graceful ferns and the gorgeous lilies, and, confined in the comfortable hut provided by charming hosts, I turned my mind to another field of study, and called Shifaladzi.

Shifaladzi is a Venda of the Balaudzi clan, a clan which has its abode at the foot of the Lomondo Hill, a kind of big volcanic cone, standing like a sentinel in front of the big chain. He had become a Christian lately and has been baptized. He has settled on the Presbyterian Mission Station at Gooldville, and he was just the kind of man you want to get sure and full information on the customs of the tribe: a grownup man, who has been a heathen up to the adult age, but who is now sufficiently delivered from superstition to explain the mysteries of his former life to a sympathetic inquirer. An important point for me was that he knew the Thonga language enough to understand my questions and to answer them in a way comprehensible to me. Thus, instead of collecting ferns, I gathered ethnographic data with the same zeal which I intended to apply to botany; and I may say, after all, the result was much more satisfactory, as I always thought documents on the primitive customs of humanity greatly exceed in value specimens of natural history!

The Ba-Venda are a very peculiar tribe, still little known to ethnographical science. The Berlin missionaries, who have been settled amongst them for nearly fifty years, have studied it and know it well, but they have not yet published a full account of its customs. One of them, the Rev. Beuster, had gathered a great many notes on the subject, but he died before having put them in order. His colleague, the Rev.