450 



Duplicate Twins and Double Monsters 



fifteen, or over 21 per cent of the whole, and in this latter computation 

 there is not included one in set VIII which happens to be a reversal, 

 and is probably a coincidence, since there seems to be in fraternal s no 

 especial tendency towards this phenomenon. Again, taking the indi- 

 vidual sets and placing over against each the number of differences they 

 exhibit, we have the column at the right hand of the table, which shows 

 very clearly the comparison between the two sorts of twins. 



TABLE IV.— SHOWING THE CORRESPONDENCE IN FINGER PATTERNS. 



A correspondence of pattern is indicated by (o), a reversal by (•) and a disparity by (x). 

 Where the material is insufficient to allow a trustworthy comparison, the space is left 

 blank. 

 No. XIII is a doubtful case. May be duplicate. (See above.) 



Again, it must be borne in mind that, exactly as in the case of the 

 palmar features, a difference which would be expressed in a formula is 

 not necessarily of much morphological importance, since here, as Galton 

 shows in spite of his desire to the contrary, patterns do merge into one 

 another by various indefinite steps, and one continually meets with a 

 pattern so near another type that its classification, even by two experts, 

 might well vary. 



This can be shown by the finger patterns of No. I, reproduced in 

 Plate B and particularly fitted to illustrate this point, since the twins 

 from which they are printed are unmistakable duplicates, and since the 

 finger patterns show, besides a typical reversal of index patterns, two 

 differences of a degree sufficient to affect the formulas. 



As nearly as I can interpret and apply the Galton system, the formu- 

 lation of these patterns, in the same arrangement as on the plate, would 



