910 llEPORT — 1900 



ever primitive in their mode of life, they are anatomically truly terrestrial aild 

 human. The present communication is only a preliminary account of Mr. Laidlaw's 

 results • moreover, there is much information available through the eBbrts of the 

 Skeat Expedition regarding the mode of life, language, customs, and religious 

 belief of these fast-disappearing aborigines. The Association is therefore to be 

 congratulated on having, by contributing to the Skeat Expedition, assisted 

 materially in rescuing these records of the Pangan tribe of the Malay Peninsula. 



7. On Crania collected by Mr. J. Stanley Gardiner in his Exjjedition to 

 Rotuma. By W. L. H. Duckworth. 



The subject of this communication is a collection of nine crania from the above- 

 mentioned locality. The results of a craniological investigation show that while 

 considerable individual differences exist, there are at least two types of skull to be 

 met -with in the island of Rotuma. These are in the first place a variety of the 

 form of cranium usually found among Polynesian natives, though possessing certain 

 characteristics which may almost be described as Mongolian ; and in the second 

 place the type of cranium characteristic of Melauesians occurs in Mr. Gardiner's 

 collection. That such different types should be met with in one small island is in 

 accordance with what would be expected on a priori grounds when it is considered 

 that Rotuma is situated at the centre of contact of three important ethnical areas, 

 viz., the Polynesian to the east, the Melanesian to the south-west, and the 

 Micronesian (where Mongolian elements are discernible among the natives) to the 

 north-west. 



8. A System of Classification of Finger Impressions. By J. G. Garsox, 

 M.D., Bxjjert Adviser and Instructor on Identification, Home Office. 



This system of classification of finger impressions has been devised to be 

 worked in conjunction with classification of records by measurements of the head 

 and limbs for the purpose of facilitating search for previous records of criminals. 

 It is also applicable for the classification of small collections of records without 

 the concurrent use of measurements. 



The objects aimed at in preparing this classification have been to get a moderate 

 number of divisions under which the various patterns and combinations of patterns 

 occurring in the arrangement of the ridges of the skin on the palmar surface of the 

 terminal phalanges of the digits of the hand may be classified, so that each divi- 

 sion shall contain approximately an equal percentage of the total number of 

 records dealt with, using for the purpose the fewest number of digits that will 

 suffice to get such a distribution, and including only the impressions of those digits 

 ■which are taken in all countries where the Bertillon system of identification has been 

 adopted — namely, the first four digits of the right hand. The patterns which the 

 ridges form are four in number, and are graphically indicated by the use of the 

 following signs :^ 



An arch thus /\ 



A loop which opens on the left . . . , / 



A loop opening to the right . . . . \ 



A whorl of any kind O 



The representations of the ridge-patterns depicted by these signs from the im« 

 pressions of the several digits in order of succession constitute the finger formula 

 of an individual, which should be noted on a prominent part of each record so as 

 to be readily seen. 



It has been found that the patterns and their combinations on the thumb and 

 three following fiugers may be conveniently arranged in ten divisions. The thumb 

 and forefinger are always required in the classification, but when the divisions 

 given by these digits are large, the middle finger or the middle and ring finger 



