mistaken hypothesis that guided Becquerel in the quest that led to his momentous discovery. Mistaken hypotheses strew the path of any energetic experimental investigator. They are as much the signs of healthy scientific life, as the carbonic acid we exhale is a sign of physical activity. For, as Sir Joseph Thomson has well said, the hypothesis, or theory, of the scientific man is not so much a creed as a policy, to be abandoned when it seems no longer likely to lead to further discovery. His articles of faith gain their validity from the logical, mathematical, and experimental checks to which he perpetually seeks to submit the conclusions deduced from them, and the only hypotheses that can ever hope to gain admission to the category of established scientific truth are those that lead to deductions which may be so tested-anywhere, at any time, by any competent observer—(quod ubique, quod semper, quod ab omnibus, to use St. Vincent's much-quoted phrase). This exclusion by the scientific man of all matters of opinion unsupported by experiment is rigorous, but it is necessary, because he feels that in this way only he is saved not merely from wasting his time, but also from that intellectual presumption which has constructed the many theologies, cosmogonies, and philosophies which, having no sounder foundation than tradition or conjecture, have, one after another, like unsubstantial pageants, vanished.

It is this humble and disciplined attitude of mind, sometimes abandoned indeed in moments of temptation by this or that individual, but sternly insisted on by scientific men as a body, that has earned for them a great historian's generous tribute.

"If men of science owe anything to us," wrote Lord Acton, "we may learn much from them that is essential. For they can show us how to test proof, how to secure fullness and soundness in induction, how to restrain and employ with safety hypothesis and analogy. It is they who hold the secret of the mysterious property of the mind by which error ministers to truth and truth irrevocably prevails."

[NOTE.-Readers who desire a more detailed account of recent researches in Radioactivity are referred to the admirable summary given in Cox's Beyond the Atom. (1913. Cambridge Manuals: price ls.) The standard English work on the subject is Rutherford's Radioactive Materials. (1913. Price 15s.)]

TWENTY-SEVENTH REPORT OF THE SCIENTIFIC

MEMORANDA COMMITTEE.

TWENTY-SEVENTH REPORT of the Committee-consisting of Mr. John S. Amery, Mr. Robert Burnard, Mr. E. A. S. Elliot, Mr. H. Montagu Evans, Mr. George M. Doe, and Mr. H. B. S. Woodhouse-for the purpose of noting the discovery or occurrence of such facts in any department of scientific inquiry, and connected with Devonshire, as it may be desirable to place on permanent record, but which may not be of sufficient importance to form the subject of separate papers.

Edited by GEORGE M. DOE, Secretary of the Committee.

(Read at Tavistock, 22nd July, 1914.)

THE Committee have pleasure in presenting their twentyseventh Report, consisting of contributions received from the following members of the Association, whose names are appended in each case: Messrs. Robert Burnard and H. B. S. Woodhouse and the Rev. J. F. Chanter.

ARCHEOLOGY.

COMBE MARTIN HALF-CROWN.

At the end of February, 1914, a half-crown, stated to have been minted at Combe Martin, was sold by Messrs. Sotheby and purchased by Messrs. Spink and Son for £13 17s. 6d. By the courtesy of Messrs. Spink I examined this coin. It is a half-crown of Charles I, dated 1645, and is of a peculiar type-no other type has supporters to the arms, and the garter round the arms has a peculiar extension with what may possibly be the monogram CM on it.

There is nothing else to connect it with Combe Martin except the tradition and the quality of the silver, which is debased and with what appears to be a small admixture

of lead. It is very similar in quality to that of some spoons made in North Devon, and also said to be of Combe Martin silver. They discolour quickly and do not readily take a polish.

I should be inclined to think it most unlikely they were minted at Combe Martin-though it is possible they were minted at Exeter from silver obtained from Combe Martin. J. F. CHANTER.

HURSTON RIDGE STONE ROW-CHAGFORD COMMON.

This fine stone row was investigated by the Dartmoor Exploration Committee when many of the stones were re-erected. Since that time several of them were again prone, and the writer determined to set them up in a more permanent fashion. This was accomplished on 21 May last with the help of Mr. George French and Mr. H. J. Roberts, the latter a new member of this Association.

Ponies are especially responsible for mishaps to standing stones. If a stone is high enough to serve as a rubbingpost it runs considerable risk, even if firmly embedded. Experience has shown that loose stones strewn around the base of each acts as a deterrent-unshod ponies fight shy of standing on such, and in every case where ponies and cattle have worked the soil away so as to form shallow pits around the standing stones these are filled with large, sharp rubble. In most cases the original reception pits were deepened, and each example dealt with was trigged in a thorough

manner.

In the south row three leaning stones were set up vertically and four fallen were re-erected.

In the north row three prostrate examples were set up. The blocking stone at the east end, and weighing more than half a ton, was inserted in a pit nearly three feet deep. It was well secured with triggers, and should now stand indefinitely. ROBERT BURNARD.

PROSTRATE MENHÎR IN BRIMPTS OUTER NEWTAKE.

This fine menhîr was thrown down many years since, and in falling broke in two. The stump, which turned out to be six feet long, remained in the ground at an acute angle. The major portion, twelve feet two inches in length, lay prostrate. The breakage was due to a fault in the

monolith-the stone being cracked vertically for some distance.

The throwing was accomplished by digging a pit at the back (north) of the stone.

It was probably intended to use up the stone for gateposts when the neighbouring newtake wall was built; the breakage and fault in the prostrate portion no doubt saved it from this appropriation.

But for its somewhat remote position, weight, and fault, it would have been used up within the past few years. It has been in danger more than once during that period. We have been anxious for some years past to re-erect the prostrate portion so as to render it safe from plunderers, and in response to a suggestion made by the Rev. S. BaringGould the Duchy Council generously offered to bear the cost. The menhîr now stands in its original position at the northern end of a much plundered stone row which, running approximately north to south, intersects the newtake wall dividing the outer and inner Brimpts newtakes.

As it was found impossible to clamp together the fallen portion with the stump in any secure fashion, it was decided to bury the latter in a horizontal position and place the remainder of the menhîr on it with its toe inserted in a space purposely left between the horizontal stone and the large vertical slab of granite which acted as a massive trigger on the south side.

The reception pit, which was somewhat deepened below its original level, was three feet eight inches in depth, and into this the monolith was placed by means of a triangle and patent blocks on 21 October, 1913, by Mr. George French, of Postbridge, under the superintendence of the writer.

Several cartloads of stones were used as triggers. These were embedded in concrete and the surface restored with herbage.

The menhîr now stands eight and a half feet above ground and is a striking landmark.-ROBERT BURNard.

ROCK SHELTER ON BELLAFORD TOR.

On 24 October, 1913, a Postbridge resident, whilst watching the Harrier Hunt on that day, observed some flint flakes under a boulder near the summit of the tor.

This was reported to the writer, who proceeded next day with Mr. George French and the finder of the flakes to the site, a narrow cleft between two large boulders. This cleft points north-west, is ten feet long, and varies in width from three feet at the entrance (south-east) to four feet four inches. On excavation, the rock floor of the shelter was found to be two and a half feet under the grass surface, with the foundation stones of a wall which evidently closed the north-west end.

Near this wall was a large fire or cooking-hole occupying a natural depression in the rock floor. This was full of peat, charcoal, and ashes.

A flint flake lay on the floor between the fire hole and the wall, and no less than sixty-six examples were found on a ledge of rock which ran under the boulder on the south side of the shelter.

This ledge formed a useful seat near the fire hole, or could have served as a shelf.

The entrance was protected by a large boulder with a passage on each side.

The shelter was easily roofed in with " vags or thatch of heather or rushes providing plenty of head room.

The earth thrown out during the excavation was composed of granite grit mixed with black soil-the latter probably derived from the roof.

No wood charcoal was observed-nor were any potsherds found.

The cooking or fire hole and flint flakes indicate prehistory. In the absence of sherds it cannot be confidently assumed that the rock-shelter goes back to so remote a time. On the other hand, it is difficult to account for the flakes and attending circumstances, unless we have in this shelter a varying instance of occupation by the hut-circle folk.

The site can easily be found by ascending the tor as far as the plateau near the summit (often used as a picnic site on Bellaford Tor Hunt day), when it lies between this and a sheer drop of several feet on the north-west slope of the tor right opposite Powder Mills.

The flints are mainly large thin flakes, mostly derived from the chalk-many show the bulb of percussion—a few only have secondary working.

The best examples are a skin scraper-a notched scraper and a borer of cherty flint.-ROBERT BURNARD.