Dec, 8, 1887] 



NATURE 



129 



as we know it in the London Basin. This will be seen on the 

 publication of a paper which was sent in to the Secretary of the 

 Geological Society on October 10 last, but has not yet been 

 put down by the Council for i^eading. A. Irving. 



Wellington College, Berks, December 2. 



The Ffynnon Beuno and Cae Gwyn Caves. 



Mr. Worthington G. Smith's letter in Nature of Decem- 

 ber I (p. 105), is so misleading that I hope I may be allowed to 

 reply to it. As is usual with highly prejudiced observers, he 

 has attempted to prove too much for his case, as he might have 

 seen had he taken the trouble to refer to my papers. The scraper 

 which he mentions was submitted to Dr. John Evans for his 

 opinion, and his conclusion as given in my paper in the Pro- 

 ceedings of the Geologists' Association, vol. ix. p. 17, is as follows. 

 The scraper " is not of a river-drift form, so far as at present 

 known, but is precisely like many from the French caves of the 

 reindeer periods, such, for instance, as La Madelaine." Mr. 

 Worthington Smith's contention therefore that it agrees exactly 

 with " the Neolithic scrapers of Icklingham and Mildenhall " 

 can only prove that there is no chronological value in the classi- 

 fication of such implements. I must explain, however, that we 

 have based no argument on the scraper referred to, since it was 

 found, before the explorations were properly commenced, in an 

 open part of the cavern, and, as stated l)y me in the paper referred 

 to, " it would be improper to dogmatize on this evidence." I 

 may say at once that I entirely demur to any classification based 

 on the form of the implements rather than on the fauna asso- 

 ciated with- them, and I see no reason whatever to suppose that 

 the worn, roughly-trimmed implements usually found ia river 

 gravels are older than the better-preserved flakes and trimmed 

 implements found in caverns, which would be used for a 

 different purpose from the rougher ones. The implements 

 discovered subsequently belong to the so-called oldest types 

 found in caverns, and were associated with Mammalian re- 

 mains, equally characteristic of the oldest river gravels as of 

 the caverns. Mr. Smith's statements in regard to the drift "in 

 front of the Denbighshire caves " are of so extraordinary a 

 character that I am tempted to ask him, before I criticize those 

 statements, whether he ever visited the FCynnon Beuno Caves 

 during the course of the explorations, whether he ever saw the 

 section of the drift exposed at the Cae Gwyn Cave, and what 

 evidence he can bring forward to support his statements that the 

 drift " is not in its original position, but distinctly and obviously 

 relaid," and that he doubts " whether before it was relaid it was 

 a true Glacial gravel at all ? " I think the members of the 

 British Association Committee, who have carefully conducted the 

 explorations, and have the strongest evidence in support of their 

 conclusion that the caverns, which are now about 4CX) feet above 

 sea-level, were occupied by man and the animals before the 

 marine drift and boulder-clay covered them over, have a right to 

 ask for the data upon which such statements as those above 

 referred to are based. These relate to facts, and must be dealt 

 with in a different manner from those statements whicli are made 

 clearly from a bias against the idea of Glacial and pre-Glacial 

 man. Mr. Smith says that he has not been able to read up the 

 literature of the subject, therefore he is probably unaware of the 

 fact that Prof. Prestwich has recently (Quart. Journ. Geol. Soc. 

 for August last) stated that he has arrived at the conclusion that 

 the high-level gravels, with implements, in the valleys of the 

 Somme, Seine, Thames, r.nd Avon date back to Glaci=il or jne- 

 Glacial times; and that "the great masses of gravel in the 

 neighbourhood of Mildenhall and Lakenheath, also containing 

 flint implements, are certainly n t of fluviatile origin" ; and that 

 they seem to him " to be part of the phenomena connected with 

 the passage of the great ice-sheet over the eastern counties, and 

 in that sense pre-Glacial." Henky Higks. 



Hendon, December 2. 



Cloud Movements in the Tropics, and Cloud 

 Classification. 



A FEW months ago I called attention to the fact that the 

 general movement of the upper clouds in the tropical regions of 

 the Atlantic was from a westerly point ; since then I have 

 worked up all my observations (which extend over a period of 



331 days spent in these regions in all months of the year except 

 June) with the following results :— 



Taking a general mean for the whole region, this gives for the 

 upper layer of clouds N. 86^° W., and for the middle layer 

 of clouds S. 73° W. These results are from observations taken 

 by myself, and no observation was registered if there was the 

 slightest doubt as to the cloud movement. The ordinary ship 

 register of upper cloud movements is worse than useless, a pro- 

 pagatory movement of the upper clouds being constantly 

 mistaken for their real movement, and the names being 

 hopelessly mixed, the cirro-cumulus being the source of most 

 mistakes. 



The cirro-cumulus exceeds all other forms of cloud in extent, 

 ranging from the delicate fine mottles at a great elevation to the 

 large flaky masses quite low down, and until it is considered a 

 middle layer cloud we are certain to have some confusion. 



It is quite time that cloud classification was placed on a more 

 satisfactory basis. Now one observer will call a certain form of 

 cirro-cumulus, a cumulo-cirrus ; a moderately high (middle layer) 

 stratus of uniform texture, a cirro-stratus ; again, one form of low 

 stratus, a pallio-stratus. Another observer will even call a de- 

 tached fragmentary stratus, cirro-cumulus ; and lots of ob- 

 servations will be useless from one observer failing to understand 

 the particular form of cloud A calls pallio-stratus or B calls 

 cirro-cumulus. Far better to keep to Luke Howard's simple 

 nomenculature till some classification is definitely fixed to which 

 all can agree. 



To be satisfactory the classification must be founded on the 

 physical and morphological (if I may use the word here) structure 

 of clouds. I find no difficulty in making observers understand 

 the difference between a stratiform and a cumuliform cloud ; this 

 is the first step, and once the distinction is thoroughly grasped 

 the rest is comparatively easy. I propose something of this sort. 

 Two orders, the " Straliforms " and the " Cumuliforms," these 

 to be subdivided into types, and these again into species ; e.g. 

 taking the ordinary dull-looking stratus commonly seen in 

 anticyclonic areas, it would be described as— 



Order ... Stratus. 

 Type ... Low-stratus. 



Species ... Pallio-stratus. 



Or take that form of cirrus which appears as lines or threads right 

 across the sky ; it would be destribed thus — 



Order ... Stratus. 

 Type ... Cirrus. 



Species ... Cirro-filum. 



By using this system an observer would be gradually brought to 

 recognize first the broad distinctions and then the minute 

 distinctions in clouds. David Wilson-Barker. 



THE FORMS OF CLOUDS. 



SO much attention has been given of late years to the 

 study of clouds, and so many names have been sug- 

 gested by different writers for the same form of cloud, 

 that the whole question of cloud foims and cloud names 

 must soon be referred to an International Congress. A 

 few remarks on certain broad facts connected with the 

 shapes of clouds, and on the fundamental principles by 

 which weather forecasts are deduced from these forms, 

 may therefore be acceptable to those who have not given 

 special attention to the subject. 



The two most important facts which must never be for- 

 gotten are : (i) that cloud forms are essentially the saine all 

 over the world ; and (2) that there are only five or six 

 distinct structures of clouds. 



