TRANSACTIONS OF THE SECTIONS. 103 



teries of some -well-marked nationality, were set before me, I would venture to say, 

 after considtation and comparison, that it might be possible to show that unassisted 

 cranioscopy, if not invariably right, even imder such favourable circumstances, was 

 nevertheless not wrong in a very large number of cases." If it is true on the one 

 hand that in gcneralibus kitet error, it is true on the other that security is given us 

 by the examination of large numbers for the accuracy and reliability of oiu" averages, 

 a principle which Gratiolet informs us is thoroughly recognized in Chinese meta- 

 physics, and which he has formulated in the foUovdng words : — "L'in variability dans 

 le milieu s'applique a tout. La verite n'est point dans un seul fait mais dans toua 

 les faits; elle est dans les moyennes, c'est-a-dire dans une suite d'abstractions 

 formulees apres le plus grand nonibre d'observations possibles." (Memoire sur 

 les Plis c^rebraux, p. 93). The natural-history sciences do not usually admit 

 of the strictness which says that an exception, so far from proving a rule, proves 

 it to be a bad one ; rather are we wise in saying that in them at least the univer- 

 sality of assertion is in an inverse ratio to that of knowledge, and that the sweep- 

 ing statements dear, as Aristotle long ago remarked (Rhetoric, ii. 21. 9 & 10 ; 

 ii. 22. 1), to a class which he contrasts with the educated, are abhorrent to the 

 mind of organic natm-e. It is true enough, as is sometimes said, that when 

 opinions and assertions are always hedged in by qualifications, the style becomes 

 embarrassed, and the meaning occasionally hard to be understood ; but this diffi- 

 culty is one which lies in the very nature of the case, and the real excellence of 

 style does not consist in its lidling the attention and relieving the memory by 

 throwing an alliterative ring on to the ear, but in the furnishing a closely fitting 

 dress to thought, and an accurate representation of actual fact. 



If we are told that the attempt to harmonize the results, not merely of cranio- 

 scopy, but of any and all natural-science investigation, with the results of literary 

 and linguistic research, is needless and even futile, this is simply equivalent to say- 

 ing that one or other of these methods is worthless. For as Truth is one, if two 

 routes purporting both alike to lead to it do not sooner or later converge and har- 

 monize, this can only be because one or other of them fails to impinge upon the 

 goal. It is true that by certain lines of investigation light is thrown upon a pro- 

 blem only at a single point, and that all fm-ther prosecution of investigation along 

 that line will but lead us ofl" at a tangent. Still the throwing of even a single 

 ray upon a dark surface is an achievement with a value of its own ; and it is a 

 cardinal rule in our sciences never to ignore the existence of seemingly contradic- 

 tory data, in whatsoever quarter they may show themselves. For what would be 

 said of an investigator of a subject such as physical geography, who should declare 

 that he would pay no attention except to a single set of data, when he was discus- 

 sing whether a particular archipelago had been formed by upheaval, or should be 

 held to be the fragments and remnants of a disrupted continent; and that if 

 geological evidence was in crying discord with his interpretation of the facts 

 of the distribution of species, it was not his business to reconcile them. He 

 woidd be held to have neglected his business, as you may see by a reference to 

 Mr. Bentham's Address to the Linnean Society, May 24, 1869 (Linn. Soc. Proc. 

 for 1869, p. xcii *). 



The argument from identity of customs and practices to identity of race is liable 

 to much the same objections and to much the same fallacies as is the argument 

 from identity of cranial conformation. The case may be found admirably stated 

 in Mr. Tjdor's work on the ' Early History of Mankind,' p. 276, ed. 2 ; and I may 

 say that the means of bringing the problem home to one's self may be found by a 

 visit to any collection of flint implements. In such a collection, as Mr. Tylor has 

 pointed out, p. 205, we are very soon impressed with the marked uniformity which 

 characterizes these implements, whether modern or thousands of years old, whe- 

 ther found on this side of the world or the other. For example, a flint arrow-head 



■* The following references to passages of the kind referred to above as to the imtrust- 

 worthiness of craniographical evidence may be useful : — Greographisches Jahrbuch, 1866, 

 p. 481. Hyi-tl, Topograph. Anatomie, i. p. 13. Henle, System. Anat. i. 198. Krause, i. 2, 

 p. 251. Archiv fiir Anthropologie, Holder, iOid. ii. 1, p. 60. See also His and Elitimeyer, and 

 Ecker in their systematic works severally, the ' Crania Helvetica ' and the ' Crania Ger- 

 maniffi meridionalis.' 



