614 



NA TURE 



[November 4, 1922 



Research Items. 



A Long Barrow in Breconshire. — In the October 

 issue of Man, Mr. C. E. Vulliamy describes the results 

 of his excavations of a long barrow at Talgarth in 

 Breconshire, on a foothill of the Black Mountain 

 range. The chamber and its contents had been dis- 

 turbed, but not in recent times. At an early stage 

 a calcined thigh-bone of a youth was found, but 

 lower down there were abundant human remains, 

 much broken and seldom lying in* anatomical relation 

 to each other, but showing no evidence of cremation. 

 Sir Arthur Keith, who has examined the fragments, 

 finds one skull of a man about forty years of age, the 

 cephalic index 70, a very narrow, relatively high, and 

 rather small head, 20 mm. narrower than previously 

 recorded in Neolithic skulls from Wales. Associated 

 with the human remains were bones of the pig, ox, 

 goat, and cat, and quantities of flint flakes and 

 scrapers are scattered over the neighbourhood. 



A Third-century Birmingham. — Preliminary ex- 

 cavations have just been completed on the site of 

 what the Times calls " a third-century Birmingham," 

 Ariconium, in the Wye valley between Monmouth 

 and Gloucester, near Weston-under-Penyard, three 

 miles from Ross. Over an area of more than 100 

 acres the earth is full of smelting refuse ; evidently 

 a great iron industry flourished there, the iron ore 

 being brought from the Forest of Dean. The only 

 classical reference to the place is in the Itinerary of 

 Antonius, compiled about 150 a.d., and in the four- 

 teenth century the Benedictine monk Richard of 

 Cirencester refers to it. Some buildings have been 

 found, the walls of which were decorated in column. A 

 large quantity of pottery, fragments of Samian ware, 

 and a coin of Domitian dated 87 a.d. were also dis- 

 covered. Ariconium seems to have arisen as a 

 halting-place on the Roman road from Caerleon to 

 Silchester in the first century, and it became a busy 

 industrial town in the third. The results of these 

 excavations justify their continuance on a wider 

 scale. 



Bacteriology of Canned Meat and Fish. — The 

 results of an investigation by Dr. W. G. Savage and 

 Messrs. R. F. Hanwicke and R. B. Calder on the 

 bacteriology of canned meat and fish have been 

 published by the Food Investigation Board as 

 Special Rep. No. 11 (H.M.S.O., price 2s. 6d. net.). 

 The report is based on the examination of 344 samples, 

 the object being to ascertain the character of bacteria 

 concerned in the spoilage of these canned foods. 

 Moulds and yeasts are of rare occurrence and probably 

 of little importance. Obligate anaerobic bacilli are 

 rarely present in sound tins, but were nearly always 

 associated with obtrusively decomposed conditions 

 in the tin. Sporing aerobic bacilli are frequent in 

 sound samples, and many of them must be regarded as 

 potential causes of decomposition ; they are unable 

 to develop in sound tins from which air is excluded 

 and persist as harmless spores. Non-sporing bacilli 

 were found in many samples, their importance depend- 

 ing upon their biological characters. Thus, strains of 

 B. proteus are important causes of decomposition. 

 Thermophilic bacteria (i.e. bacteria growing best at 

 55 C.) were searched for and found to be widely 

 prevalent but, being non-proteolytic, are unlikely to 

 cause spoilage. Micrococci are infrequent and as a 

 group cannot be regarded as a cause of spoilage, 

 though they may assist more proteolytic types of 

 organisms. Nearly 62 per cent, of sound tins are not 

 sterile, the worst offenders being crab and lobster. 

 Sterility itself is therefore not a criterion of sound- 



NO. 2766, VOL. I IO] 



ness, and these surviving bacteria do not in any way 

 injure the foods in which they are present owing to 

 their inability to multiply and produce decom- 

 position under the conditions existing. Suggestions 

 are made respecting the process of manufacture so as 

 to reduce spoilage to a minimum. 



Flower Structure in the Lecythidaceje. — 

 Prof. McLean Thompson has published a further 

 study of floral morphology in the Lecythidaceae, a 

 peculiar tropical family with large fruits, related to 

 the Myrtaceae (Trans. Roy. Soc. Edin., Vol. 53, 

 Part I., No. 13). The present paper is devoted to a 

 study of the flowering and certain stages in the floral 

 development of Napoleona imperialis, the first 

 member of the West African genus Napoleona to be 

 described, in 1786, by a French writer. The peculiar- 

 ities of the flower include so-called inner and outer 

 corollas with a series of petaloid filaments between 

 them. Many interpretations have been placed upon 

 these structures. The floral development indicates 

 that the petaloid filaments and the outer corona have 

 taken the place of the outer cycles of stamens in the 

 Myrtaceous flower, as Bentham supposed. The style 

 is surrounded by a fleshy glandular disc which is 

 considered to be a remnant of an inner stamen- 

 bearing whorl. 



Hawaiian Grasses. — A comprehensive account of 

 the grasses of the Hawaii islands, which is based on 

 special collections made by the author in 1906, and 

 on all accessible material gathered by other observers, 

 is provided in Memoirs of the Bernice Pauahi Bishop 

 Museum, Vol. VIII. No. 3 (Honolulu), " The 

 Grasses of Hawaii," by A. S. Hitchcock. The 

 preliminary discussion deals with the distribution 

 from an ecological standpoint, with the chief agri- 

 tural grasses, and with introduced species. A large 

 number of the grasses of the islands have been 

 introduced, 83 species in all, mostly from Europe, a 

 few from Australia and the East Indies. Many of 

 the 47 native species are endemic, and most of the 

 others have extended northwards to Hawaii from the 

 East Indies and the southern Polynesian islands. Of 

 the 39 endemic species 7 are annuals, all belonging 

 to the genus Panicum. The second part is systematic 

 and gives a full description of each species, the 

 necessary keys being provided ; 9 new species are 

 described, together with a new section of Poa, Siphono- 

 coleus sect. nov. The treatise should provide a useful 

 and well-arranged reference work for students of the 

 Graminaceae. 



West Indian Hurricanes. — An article on the 

 formation and movement of West Indian hurricanes 

 by Mr. E. H. Bowie of the U.S. Weather Bureau 

 appears in the U.S. Monthly Weather Review, April 

 1922. The area of first appearance is described as 

 extending from latitude 10° to 25 N. and from 

 longitude 56 to 95° W. During a period of 35 

 years, go per cent, of the West Indian hurricanes 

 are said to have had their origin within and not 

 without this area. It is asserted that some years 

 pa'-s without hurricane formation, while other years 

 are notable for hurricane frequency. The origin of 

 a hurricane is by no means certain, and the author 

 states it is even now difficult and next to impos- 

 sible to say which of the many hypotheses is the 

 correct one. The prevalent supposition of the origin 

 is the meeting of adverse currents, having different 

 temperatures, which produce gyratory motions of 

 the atmosphere. Many men of science are of opinion 



