Maech 25, 1887.] 



SCIEJSrCE, 



295 



set an example which is being eagerly followed 

 by the historical bodies of most European coun- 

 tries. Germany, indeed, is, as well as America, 

 already ahead of us in scientific methods of col- 

 lecting and editing the more modern and political 

 materials vvliich may be gleaned from the archives 

 of every state-paper office in Europe ; while 

 France, Austria, Belgium, and Sweden tread 

 closely on our heels. The objects of modern his- 

 tory, therefore, though professedly national, are 

 in fact cosmopolitan, each country opening up at 

 times unexpected manuscript treasures for the 

 more particular advantage of the other. Hith- 

 erto we have been content to rely chiefly upon the 

 resources of our unrivalled national records ; but 

 every year affords fresh evidence of the exteiit 

 and value of the outlying manuscript material 

 which it is the special mission of the Historical 

 manusci'ipts commission to incorporate with the 

 main stock." 



— Our retinal insensibility to the ultra-violet 

 and infra-red rays has been recently discussed by 

 Drs. Fox and Gould in the American journal of 

 ophthalmology. The sufficient reason for the per- 

 ception of the so-called 'light' rays is because the 

 eye has learned to react to the strongest and most 

 constant stimulus, and to extinguish or exclude 

 those vibrations that would only confuse by their 

 weakness or inconstancy, or that would with 

 difficulty be focused with the rest. As to the 

 range of vision along the spectrum, the remark- 

 able fact is, not its narrow limits, but its extension. 

 The marvel is that we have learned to see the 

 violet rays at all, when they are so weak. The 

 limit at the red end of the series is thought to be 

 determined by the great absorption gap in the 

 spectrum that separates the visible from the infra- 

 red rays. It is then asked, how are the invisible 

 rays excluded from stimulating the nerves ? and 

 although no satisfactory or final answer can be 

 given, based on experiment, it is made at least 

 probable that they are absorbed by the media of 

 the eye before they reach the retina. 



LETTERS TO THE EDITOB. 



*t*CorresiJondents are requested to be as brief as possible. The 

 writer^s name is in all cases required as proof of good faith. 



A sensitive wind-vane. 



An interesting discussion of this question has re- 

 cently been initiated, and it may be well, to give a 

 portion of this and a few considerations bearing on 

 the problem. I have seen it stated that a flat vane 

 is always in a neutral li^ie, and a sensitive one is made 

 by fastening two plates together at an angle of about 

 ten degrees. This statement has always appeared 

 chimerical to me, for the reason that such a vane as 

 described would have twice the weight and friction of 

 a flat vane, and hence could not be as sensitive as the 

 latter. "We should gain, at the outset, a clear defini- 



tion of what is meant by a sensitive vane. A very 

 light structure, like a feather attached to a cord or 

 balanced near one end, while tossed hither and yon 

 by every breath, and exceedingly sensitive, could 

 hardly be what is meant. I would say, as a first 

 idea, that a sensitive vane is one that most readily 

 assumes the wind-direction. 



Professor Ferrel has discussed this question, from 

 a mathematical stand-point, in the February number 

 of the American meteorological journal. He assumes 

 that the gyratory force (gy) of the Avind upon a 

 double-tailed vane varies as the square of the sine of 

 one-half the angle between the tails, and gives the fol- 

 lowing expressions for the gyratory force. Let i = 

 one-half the angle of tails, e = angle of deviation of 

 wind, and F= wind-force upon unit surface of vane : 

 then we shall have, with i > e, gy = F sin 2i sin 2e ; 

 with i<e, gy = F sin- (i + e) in the case of a 

 double-tailed vane, and gy = F sin"^ e with a flat 

 vane. Professor Ferrel finds, that, with2z=90°, 

 there is a maximum sensitiveness of the vane. With- 

 out entering upon a discussion of the theory devel- 

 oped by Professor Ferrel, it may be suggested that 

 we cannot neglect the great pressure that the tails at 

 an angle of 90*^ would have to bear in a high wind, 

 and which would come upon the axis. This amounts 

 to ten pounds per square foot in a wind, forty miles 

 per hour, impinging normally upon a surface. The 

 angle of the sides being 45°, the total pressure would 

 be somewhat less, but would still be sufficient to 

 prevent all free action of the vane. 



Mr. G. E. Curtis has also very recently given a 

 theoretical discussion of the question before the 

 Washington philosophical society, and in this he 

 diifers very materially from the one just given. He 

 assumes that the action of the wind varies as the 

 sine of its deviation angle. He gives for vane with 

 double tails, gy = F sin (i 4- e) when i > e, and gy 

 = 2Fsin i cos e when i <e ; for a flat vane, gy — 

 F sin i. In the original formula F is omitted ; but I 

 have supplied it, as it seems necessary. The nota- 

 tion is the same as in the previous case. There 

 is a remarkable variance in these theoretical re- 

 sults, and it is a little difficult to state which is the 

 more satisfactory. I hardly think that either can be 

 accepted by the working meteorologist ; but prob- 

 ably Professor Ferrel's is the more satisfactory, cer- 

 tainly for light winds. 



No attention is paid in either of these discussions 

 to the weight or friction of the vanes, yet it would 

 seem as though either one of these is a far more 

 important element than a single or double tail. In 

 the discussion by Mr. Curtis we may very readily 

 take these factors into account by placing the two 

 tails of his double-tailed vane one above the other, 

 edge to edge. We now have a flat vane whose 

 weight, friction, and all other essentials are the 

 same practically as those of the double-tailed vane ; 

 in fact, simply a transformation of the latter, without 

 alteration except in the matter of surface. In fact, 

 both vanes are directly comparable, while they were 

 not before. We have, however, just doubled the 

 surface of the flat vane, so that gy = 2F sin i. Now, 

 it is very easy to see that this expression has a 

 greater value than F sin (i + e) when i> e, and also 

 greater than 2F sin i cos e when i < e. This theo- 

 retical discussion, then, by Mr. Curtis, shows conclu- 

 sively that the flat vane is the more sensitive. When 

 we consider that Professor Ferrel regards the flat 

 vane as much the steadier of the two, also that the 



