322 



Life and Letters of Francis (hiUon 



white wafers attaclied by paste'. Many detailed suggestions for taking the 

 |)hotof;nij»h8 are given. Galton indicates as in his earlier paper that if all 

 the animal's h(H)fs fall in the middle two feet of the path, a slight obliquity 

 to the wall will not introduce an error of any importance into the pnoto- 

 gniphic measurements (.see our p. 3'JO). (Jalton's procedure is now somewhat 

 simpler than that which he gave in Nature. He projects from the optical 

 centre on to the wall behind the horse, takes the mejisurement there and 

 reduces in the ratio of distance from that centre to median plane of horse to 

 distance of that centre from the wall. He now considers it adetpiate instejul 

 of determining the median plane by bisecting lines between tips of fore- and 

 hind-hoofs, to take the median plane 6 inches liehind the line joining corre- 

 sponding fore- and hind-hoofs. He gives reasons for believing that the 

 errors ot measurements made on the photographs are less than those made 

 by diflferent persons on the same animal. 



In the cotn-se of the paper he refers to standard fjhotographs made of 

 28 premium stallions at the Royal Agricultural Hall and to photogra]»hs 

 of 31 triads made on pure-bred shorthorn cattle, chiefly at Alnwick Park. 

 The Royal Commission on Horse-Breeding was iisked to permit a trial 

 inst-illation at their show at the lloyal Agricultin-al Hall in 1H98, and per- 

 mission was cordially granted. The installation was made and 35 horses were 

 photographed in 3^ hour.s. The total cost including that of two veterinary 

 measurers was under £25, and Galton believed that if the operations became 

 customary, the cost would be paid by the sales of copies of the standard 

 photographs. 



Galtou's Report occupies pp. 12-16 (with plate of figures) of the 7th 

 Annual Report of the Royal Commission on Ho^'se- Breeding (C. 9487, 1899). 

 He gives some strong arguments for standardise<l j)hot<»grapliy and descril)es 

 again in <letiiil the standardisetl methods, and the divergences between the 

 pnotographic and veterinary measures. He considers the latter as far more 

 reliable tnan the former. In the matter of 



"length of bixly, the ptKitographic tnotluKl is the only une to be depended on, and it seems to 

 lie as trustworthy as that of height." (p. 15.) 



Putting aside the length' we have for the horse " Maroni " for example : 



Height in inches. 



as an illustration of the extent of agreement. 



' Associated with this section of the material in the Galtonutna is a careful drawing of a 

 hor«e by W. t'. H. Weldon with anatomical sketches of the skeleUui indicating the points 

 which could In; ap))roxiniatc^d to on the living animal, and suggesting suitable mwisurements. 



* "A direct and trustworthy measurement of the length of a wioos or timid horse is 

 extremely difficult, perhaps impracticable." (p. 15.) 



