106 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



the oi^eiiiug in tbe third frame. In preparing for the rear view we take sight on the horizontal plane 

 and subnasal point from behind, and on the maximum occipital point from before. With the side 

 view the facial portion of the skull is turned toward the left. We take sight on the horizontal 

 jjlane from a position to the left of the center; on the left supra auricular point from before, and 

 on the right supra-auricular point from beliind. With the described apparatus views of the base 

 and vertex have not yet been attempted, but it is believed that lateral frames with the usual 

 cross-threads nuist be added to secure good views of these asi)ects of the skull. 



The duration of each fractional exposure depends on many conditions; the sensitiveness of 

 the plate, the refractive power of the lens, the orifice of the diaj)hragni, the degree of light, the 

 color of the skull, and the number of skulls in each series. In a series of five skulls, other things 

 being equal, each fractional exposure will be twice as long as in a series of ten. In photographs, 

 19 et seq., we used " Carbutt's Keystone dry plates," a Dallmeyer triplet, 4i-inch lens, a diaphragm 

 with 14 inch aperture, and an exposure of from 10 to 20 seconds for each skull. 



Different method-j of determining the adjustment of the skulls have been tried: First. Two sets 

 of cross threads have been used, and the operator taking sight only from the front, with his head 

 placed immediately in front of the lens. Second. Two sets of cross lines employed, the operator 

 looking throiigii the camera only, the plate necessarily removed from the camera before each exposure. 

 Third. Two sets of lines as before; an accessory camera used at the side; the front vertical thread 

 aligned on the more distant margin of the anterior nasal orifice. Fourth. Two sets of lines used; 

 a sketch of the first skull drawn on a cross-lined gelatine film, and each subsequent skull made to 

 conform as nearly as possible to this sketch. Fifth. The plan already described at length and at 

 present adopted by us, in which there are four sets of lines — one ou the ground-glass plate — and 

 which the operator views not only from the front, but from behind, through the opening in the 

 posterior frame, in order to secure a proper alignment of the maximum occipital point in the front 

 view, and of the subnasal point and the horizontal plane in the rear view. 



The apparatus used, which is illustrated in the plates, is rudely improvised from material at 

 hand. The frames were not made on purpose, but were such as we had in the museum for other 

 uses. A more convenient apparatus is to be constructed, but the general principles of the one 

 now in use will be preserved. 



Our craniophore, however well it may be adapted for the purpose for which it was originally 

 intended, is not well suited for adjusting skulls in photography. The modification of this, recom- 

 mended by Ranke figured in "Archiv fiir Anthropologic, 1S83," would undoubtedly do better, but 

 a still more suitable craniophore can, we believe, be devised, and we propose to have such an 

 instrument constructed. 



Two sets of composite photographs of crania are shown herewith, viz : One set including six 

 male Sandwich Islanders' skulls, and one set including six male Arapahoe Indian skulls. There 

 are six photograjihs in each set, and all are exactly half the size of the original objects. 



The value of this method of composite photograi)hy, as applied to crauiological studies, de- 

 pends, to a very considerable extent, upon the adoption of some uniform standard of size for the 

 preparation of such i)hotographs, in order that the series of specimeus in different uuiseums and 

 collections may be directly compared. It appears to me that the most convenient scale for such 

 photographs is to make them of one-half their natural size — that is, so that the inch divisions on 

 the graduated rule, which is always photograjihed with each set, shall measure exactly one-half 

 inch in the photograpli. 



These composite photographs must be studied in connection with the measurements of the 

 crania represented in them. The method is simply a rapid and convenient means of obtaining a 

 graphic representation of a series of irregular objects, a jdccure whicli should indicate not only the 

 mean size and shape of these objects, but also, to a certain extent, the maxima and minima of 

 their variations. I think it is much better to make the photograi)hs directly from the skulls them- 

 selves, than to construct them from separate photographic i)rints of each skull ; which appears to 

 be the method pursued by Dr. Thomson, of Edinburgh, in the specimens given by liim in the 

 "Journal of Anatomy and Physiology," London, 1885, Volume XIX, Ft. II, page 230. 



It is well known to ethnologists that the distinctions of race are much more marked in the 

 physiognomy of the living subject than in the differences shown by dried crania ; and that the 



