L. Rogers
1920 BMJ (Clinical Research Edition)  
of confirmiiation than in seeking-for exceptions whose existence would point to somiie fallacy in my proof. No such exceptions were found, but muclh valuable and interesting confirmation waas fortlhcomina. Efect of these Winds on Phthisis Patients.5-- Cases of plltlhisis in Devonshire exposed to these winds have been slhown to run a shorter course tlhan cases slheltered fromn them. Unpublislhed cliniical evidence also lias accumulated pointincg strongly in tlle same direction. I lhave also
more » ... . I lhave also tried to compare the statistics from variously placed sanatoriums ; this proved imiipossible, but it was signiticant that some exceptionally exposed sanatoriums kept no records of results! Tlhere seemns to lue no doubt that some at least of tlle disappointment generally expressed witlh the results of sanatoritum treatmiient in this country is due to the ill-clhosen sites of some of our sanatoriums. Perhaps n'w that an able indepenident investigator lhas conifirmed my conclusions, a little paiins may be taliken to place tlle new sanatoriums for our plhtliisical sailors and soldier's in situations vwhere failure s 'all be less of a foregione conclu.sion! 3. Effect of these IWinds on Bovine Tntbercztlo-sis.-Bovine tuberculosis is muuclh commoner in areas exposed to strong rain-bearing winds tllan in areas slheltereed from tlhem.6 Professor Delepine's valuable paper 7 on tuberculosis in tlle farms supplying Manclhester witlh milk gave me a unique opportunity of investigating tllis important point. This fact surely calls for sonme attention. 4. Altitude and Phthi8sis.R VWhen tlle influence of rainbearing winds had been talken into account and, as far as possible, elimi:inated by considering onlv localities slheltered fromn tlhem, the contradictions, which had rendered a satisfactory conclusion on the relations of altitude and plhthisis prevalence impossible, disappeared, and the impportant generalization was reaclhed that altitLide, per se, has no influenice on plhtlhisis prevalence. Thie fact is tllat, as altitude increases, exposed situations increase so muchi in bleakniess that the population is nmore and miiore forced into slhelter; consequently, as altitude increases tlle proportion of the population living in slhelter increases and tlle prevalence of plltllisis falls. Tllis concluLsion does not affect nmy belief in the beneficial effect of hligh altitudes in shelter on phtbisis patients. 5. Subsoil anld Phthisis.-Dr. Brownlee must hava giveni my paper.2 on tllis subject a very hasty glance, or he would hiave recognized that n-ot only did it deal with a miiuchl muore extensive field than Buclhanan's-all England as against thlree counties-but that it was much more elaborate and safeguarded from error; also (most important of all) tlhat instead of leaving the question one of mere. surmise, mine left it practically proved that certain geological formations are definitely associated witlh hiiglher plithisis death rates than others. T say "practically" proved because, until larger populations in pronounced slhelter from rain-hearing winds lhave been dealt witlh, I shall not consider the investigation conmplete. Dr. Brownlee's results in Norfoll, Suffollk, and Essex are very suggestive, but, in view of the varying exposures of hiis districts, must b3 regarded as inconclusive. 6. Occupation anzd Phthisis.-I wrote in 1913 10: "1 The evil eiminenlce of the tin-miner may not be entirely occupational; he lives for the most part in a district swept by strong rainy winds. Also tlhe Midland coal-miner, wlho suffers less from phtlhisis than coal-miiners in other parts of England, mnay not owe his good fortune altogetlher to the conditions of hlis work, but partly also to the slighter exposture of his district to rain-bearingy winds." 7. Density of Population and Phthisis.11-I lhave slhown that in certain rainy wind-swept regions the self-protection of a town from winds may mlore than compensate for its comparative crowding. PHTIISIS IN WALES. Dr. Brownlee's section on this subject is instructive and interesting. Bult I fear lhe will finid the problem of subsoil in Wales as disappointina a hie found hiis Welslh study of rainy winds. I lhave tried it myself, for I observed how heavy-were th4e deathl rates on some of thIe WVelsh formyations. Rainy winlds, race, modes of li-fe and occupation, seem perversely arranged so as to confuse thle problemn and almost to defy elimination. It will b3 observed thlat Dr. Brownlee's mnap of thle districts is almost as suggestive of the effect of rain-bearing wind as of subsoil, whlen it is remembered that in the east anid soutl-east of Wales the mountains afford much shelter, and tlhat, in the soutlh-east, occupation must mnodify the figures, the plhtliisis deatlh rate of coal-niners generally tending to be low. My table of the phtllisis deatlh rates in the Welslh registration districts, 1881-90,11 indicates at least some effect pLroduced by rainbearing winds. A MISAPPREHENSION. Before concludina, I slhould like to clear up a misapprehension. Dr. Brownlee lhas referred to me as a "follower in Haviland's footsteps." So far fromii beina a follower of Haviland, I was fortunately quIite unaware of hiis vork until I lhad founid the fact on whlich I lay such] stress. HIaviland's statenments were quite differenit fromll mine, and, lhad I been familiar with tlhem, it is not unlikely tllat I slhould have missed my point, as hle did. S UMIMIARY. 1. Dr. Brownlee has introduced a new mietlhod of inivestigating plhthisis death rates at successive age periodls whliichl promises to be valuable. 2. Hle lhas not proved hiis first proposition reaarding thle existence of two or tllree types of humiian tubercle. 3. He lhas hiimself disproved hiis second proposition, dealing with the cause of tlle decline of plhtlhisis. 4. His conifirmation of my conclusions in Devonslhire in respect of the influence of strong prevalent rain-bearina winids on tlho prevalence of phthisis, once mnore einplasizes the necessity for recognizing this influence. 5. Incidentally hiis papers illustrate tlle need for applying, in clinatoloaical research, what I have calledltlle principle of approximate isolation of influences." REFERENCES. 1An Ivestigationt into the Epidemniology of PlWtimisis int Great
doi:10.1136/bmj.1.3096.596 fatcat:np2ditdupvfipe23o7z7xkqadm