SCIENCE. 



[Vol. XI. No. 274 



ceive them both in the same plane, representing the experiment of 

 Fig. I, only with a greater number of circles. It is quite possible 

 that very manv experimenters cannot obtain the effect at all. In 

 our own case we are much aided by the readiness with which the 

 innervation of each eye can be carried on independently of the other. 

 In fact the phenomenon may be peculiar to our own experience 

 alone, and may not be capable of verification by others. We shall 

 be glad to know if it can be verii^ed. J. H. Hyslop. 



Baltimore, Md., April 26. 



Is the Rainfall increasing on the Plains ? 



I NOTICE a letter from Mr. Curtis in Science of April 20, calling 

 attention to an error in the accepted Fort Leavenworth precipitation, 

 due to the reckoning of snow (unmelted) as rain. I have looked up 

 Schotfs oricrinal manuscript, and find the large precipitation in 

 January 1871, entered "11.25"?" showing that the compiler was 

 aware of a possible error. A careful examination of the original 

 record shows that the true value is 1.20 for January, and 46.70 for 

 the year For 1872 the amount should be 51.65. I am inclined to 

 think that Mr. Curtis is altogether too sweeping in his criticism. 

 The probability of such an error having crept into the bulk of the 

 Fort Leavenworth records is exceedingly small ; and, moreover, the 

 records nearly all the way through are partially checked by parallel 

 records at neighboring stations. . , • ^u 



It is certainly true, that, " if such errors as these exist in the 

 records it is surprising to find that the rainfall of Kansas is increas- 

 in<^ " In this quotation from the letter, I have omitted a ' not be- 

 fore ' surprising.' The reason is plain. Since 1873 the Fort Leav- 

 enworth records are not quoted, but only those of the Signal Ser- 

 vice Now it is recognized that the exposure of the latter gauge 

 wiu'give too little rainfall; and, moreover, the measurement of 

 melted snow is invariably too small. Both of these causes combine 

 to render the records too small since 1873; and, if we assume that 

 before then the records averaged five to ten inches too great, it is 

 easy to see that there has been an enormous increase in rainfall, if 

 the last fifteen years average more than the previous fifteen. 



I wish to call attention to an exceedingly interesting point m this 

 connection. During the last four years. Dr. Carpenter, at West 

 Leavenworth, has reported from five to twenty-five inches more 

 rain each year than the Signal Service two miles due east. Will 

 not some scientist residing near Leavenworth take a special interest, 

 and determine the possibility of such a large increase m precipita- 

 tion in so short a distance ? This will also have a most important 

 bearing on the rainfall question. H. A. Hazen. 



Washington, DC, April 21. 



Chloride of Nitrogen. 



It seems to me worth while to call attention to the fact that the 

 preparation of chloride of nitrogen by the electrolysis of a solution 

 of ammonium chloride, announced in your last issue (p. 206) as a 

 new discovery, has, as a matter of fact, long been known. This 

 method of preparation forms, indeed, one of the stock lecture- 

 experiments in many courses in chemistry. Incidentally it may be 

 noted that within the last few weeks chloride of nitrogen has been 

 made in considerable quantity in Gottingen by Dr. L. Gattermann, 

 who has also for the first time made careful analyses of the sub- 

 stance. The difficulty involved in such an investigation will be 

 appreciated to some extent when it is borne in mind that chloride of 

 nitrogen is probably the most explosive chemical compound known. 

 Dr. Gattermann's investigation has been spoken of in German news- 

 papers as an act of heroism. IRA Remsen. 



Baltimore, April 28. 



Christmas Customs in Newrfoundland. 



In Science for Feb. 24, 1888, it is said, in the note on ' Christ- 

 mas Customs in Newfoundland,' that the practice there described 

 of tying a wren to a bush, and singing the rhymes there given, is 

 not known in other places. It may interest your readers to know 

 that fifteen years ago certainly, and probably at the present time, 

 the country boys in County Clare and County Limerick, Ireland, 

 if not in other counties, never let St. Stephen's Day pass without 

 bringing round from house to house a bush adorned with ribbons. 



with on the top a struggling wren, or, if not a wren, some small bird 

 for that day dignified by the name. The rhymes sung during the 

 cruel ceremony were, I think, identical with those given in your 

 paper. And in some way or other the coppers which the youths 

 pocketed — given them at the houses they visited, whether on con- 

 dition of releasing the wren or not — were supposed to do honor to 

 the dying bird. W. F. StoCKLEY. 



Fredericton, Canada, April 19. 



Osteological Notes. 



It is often extremely difficult to determj^ne with accuracy the 

 boundaries of the four centres of ossification which .characterize the 

 occipital segment of the cranium in the various orders of the mam- 

 malia. The tendency to early co-ossification of these separate cen- 

 tres or bones, as they are generally described by anatomists, is for 

 the most part so great, that it is impossible to obtain the informa- 

 tion desired without the aid afforded by the collections of large 

 museums ; and, even with this advantage, perfect accuracy of de- 

 scription is scarcely possible in many cases, on account of the lack 

 of material. 



As a general rule, with notable exceptions, however, the four 

 bones — viz., the supra-occipital, two ex-occipitals, and basi-occipi- 

 tal — individually contribute, in a greater or less degree, to the for- 

 mation of th.t foramen magnum, the amount thus contributed by 

 each depending very much upon the shape assumed by that open- 

 ing (compare Figs, i and 4). 



In both the odd-hoofed and pair-hoofed animals {Ungula/a), in 

 the elephants (Proboscidia), dugong and manatee {Sirenia), in 

 the pangolins {Edentata), and in the opossum {Marsupialid), the 

 ex-occipitals meet above, and thus shut out entirely the supra- 

 occipital from participation in the margin of the foramen (Fig. 2). 



In the remaining orders it may be said that the supra-occipital 

 contributes from a third to a fourth of the border of the great open- 

 ing, the lines of suture between this bone and the ex-occipitals run- 

 ning slightly upward and outward to a point corresponding with the 

 level of the zygomatic process of the squamosal. 



The ex-occipitals, with few exceptions, as in the dog (Carnivord) 

 and in the armadillos {Edentata), supply the greater portion of the 

 condyloid surface, the remainder being furnished by the basi-occipi- 

 tal. 



The lines of suture which mark the separation of these two seg- 

 ments should be drawn from the margin of \.\\^ foramen magnum 

 downwards and outwards, bisecting the inner third of the condyle, 

 to a point corresponding with the centre of the tympanic or auditory 

 bulla (Fig. 4). In the cases where the condyles are the product of 

 the ex-occipitals alone, as in the dog, the lines of separation must 

 be drawn in the same general direction, but not so as to include 

 any of the condyloid surface (Fig. 3). D. D. Slade. 



Museum of Comparative Zoblogj', Cambridge, Mass., 

 March 30. 



