■224 



M.D., assistant physician Northern Michigan Asylum ; " The 

 Story of the Bacteria and their Relations to Health and Disease," 

 by T. Mitchell Prudden, M.D., author of "A Manual of Practical 

 Normal Histology ; " and " Through the Ivory Gate," being studies 

 in psychology and history, by William W. Ireland, author of " The 

 Blot on the Brain." 



Ex-Postmaster-General Thomas L. James has prepared an 



explanation of needed postal reforms, which will appear in the 

 October Forum. Mr. James declares that the railway mail service 

 is twenty years behind the times, and ought to be very greatly im- 

 proved ; that small offices near to one another ought to be con- 

 solidated under one management, so as to save expense ; and that 

 •ocean postage ought greatly to be cheapened. Senator Cullom of 

 Illinois will have an article in the same number on " Protection and 

 the Farmer," to show that the farmers are benefited by a protec- 

 tive system more than any other class. Mr. Edward Wakefield, a 

 member of the Australian Parliament, who has been elected and 

 defeated many times under the Australian ballot system, will con- 

 tribute to this number an explanation of the practical workings 

 and of some defects of the system which has been so much dis- 

 cussed in this country. Professor William T. Harris, United States 

 commissioner of education, writes a critical examination of Edward 

 Bellamy's " Looking Backward." 



SCIENCE. 



[Vol. XIV. No. 347 



LETTERS TO THE EDITOR. 



A New Method for ascertaining Heights and Distances in 



Right-angled Triangles. 



About four years ago I devised a method whereby thq solution 

 ■ of right-angled triangles, for the taking of distances and heights, is 

 much facilitated by a tangent scale on the instrument. 



The principle depends upon the well-known fact that the per- 

 pendicular of a right-angled triangle is equal to the tangent of the 

 ■included angle multiplied by the base. 



The graduation is accomplished as follows : we take a base-line 

 .(say, of 100, for convenience), and an angle of five degrees. Com- 



_Jvi;and DB represent the rights of an ordinary surveying-campass ; DB containing 

 the scale, and sliding upon ^5C, which contains the numbers 1-230 marked in 

 equal divisions. 



puting this, we find the perpendicular to be 8.75 feet, yards, or 

 metres, in whatever system the base was measured. 



This is marked on the arm BD instead of five degrees. The 



■ computation is continued for the various angles, and the results 

 •marked upon the scale. This for a base of 100. Now, if the ob- 

 server is placed only 50 distant from the object, DB is moved to 

 that point on the scale ABC, and the height is seen to be the same 

 as before; for, at a distance of 50, an angle of ten degrees, which 

 is observed by going one-half nearer, is subtended by a perpendic- 

 ular of 8.75, as before: so by moving the scale backward or for- 

 ward, corresponding to the base-line taken, the height of an object 

 can be immediately read off, provided the side of the object con- 

 tains the height ; if it does not, other means of triangulation have to 

 be adopted, several methods of which can be readily improvised 

 by one accustomed to such work. Horizontal angles can be solved 

 in the same manner by having the rim of the compass-box gradu- 

 ated for a given base-line ; then by using this base-line, and taking 

 the distance between the observed points to represent the perpen- 

 dicular of the triangle, the distance can be read directly from the 



.instrument. Harvey B. Bashore. 



West Fairview, Penn., Sept. 13. 



Brocken Spectre. 



This phenomenon has been associated with the Brocken, one of 

 the Hartz Monntains in Germany, about 3,700 feet in height, be- 

 cause more often observed from there. It has given rise to a large 

 number of remarkable theories in explanation, many of which 

 originated with those who had never seen it. An exhaustive arti- 

 cle, giving a resunii of records regarding it, will be found in the 

 Qicarterly Journal of the Royal Meteorological Society for 1887, 

 at p. 245. The explanation having the widest acceptance was pub- 

 lished in the above article, and later in the American Meteorologi- 

 cal Journal, August, 1 889, and is as follows : the eye is deceived 

 by the apparition, and thinks it much farther away than it really is. 

 It seems to me that this is hardly tenable. The only way in which 

 the eye could be deceived would be in case the shadow were 

 formed a long way off ; but, if it were really formed near the eye, it 

 would appear in its natural size. When one looks into a concave 

 mirror, the eye is at first deceived, thinking the mirror plane ; but 

 in this case the deception is very plainly due to the action of the 

 mirror. 



The very singular explanation is given in " Johnson's Cyclopse- 

 dia," that " the vapors of the atmosphere act as a vast concave 

 mirror." Singular as it may seem, however, it is probable that this 

 is, undesignedly, more than half correct. A short stay on the 

 summit of Mount Washington has shown this spectre in all its 

 phases. The best time to see it is either in the early morning or 

 just before sunset, and when the fog is not too dense to hide the 

 sun. If the observer turns his back to the sun, he will see- on a 

 bank of fog, if it does not envelop him, a slightly diminished shadow 

 of himself. The eye is not deceived in any case as long as the fog 

 forms a nearly vertical wall at fifty or more feet distance. If, now, 

 the fog envelops the person, the shadow appears to start directly 

 from him, and often seems very large. There is no deception of 

 the eye at all, if one is accustomed to careful observations. 



The following is advanced as a probable explanation. The 

 shadow of the person is cast upon the fog in solid form ; that is, 

 the object shuts off the light of the sun, and one sees only the sur- 

 face of his own solid shadow looking into its axis. The arms and 

 legs also cast solid shadows, and the person sees the movement of 

 these outside of the shadow of his body. It may be better under- 

 stood to call to mind the shadow one sees on the ground as the 

 sun is setting. This gradually grows longer and longer, and at 

 last disappears in the distance. The fog forms a sort of " ground," 

 and the shadow is cast upon it. It is possible to form the same 

 shadow with a lantern which concentrates its rays by a reflector. 

 There is no difficulty, in a fog, in seeing the shadow enormously 

 enlarged. Scores have seen it on Mount Washington. It might 

 be thought that the nearness of the light was the cause of the en- 

 largement ; but this was not the case, for the shadow began ex- 

 actly at the person where it could have been only the natural size. 



The familiar appearance of " sun drawing water " will help to 

 explain this phenomenon. In this case the air is full of haze or 

 fog, and a small cloud casts a solid shadow thousands of times as 

 long as itself. The surface of this is what we see. If an eye were 

 placed in the edge of the cloud casting the shadow, the latter would 

 appear on all sides. In the case of the spectre, this same solid 

 shadow could be seen by a second person standing and looking 

 across it, provided the light of the sun were not dimmed by the 

 fog. It is to be hoped that we may have more observations of this 

 interesting phenomenon. H. A. Hazen. 



Washington, Sept. 23. 



Note on the Anserine A£Snities of the Flamingoes. 



A CLOSER study of the structure of a member of the groups of 

 existing birds is throwing a new light in upon their relationships, 

 and at the same time somewhat disturbing some very crude and 

 preconceived notions as to their affinities. 



For a great many years past, some of the most distinguished of 

 zoologists have insisted that the position of the flamingo was " so 

 completely intermediate between the anserine birds on the one side, 

 and the storks and herons on the other, that it can be ranged with 

 neither of these groups, but must stand as the type of a division 

 by itself." Recently, Professor Parker {Ibis, April, 1889) has said, 

 in reviewing the structure of the wing in the flamingo {Pkcenicopie- 



