March 11, 1887.] 



SCIENCE. 



233 



Eoot while he was principal of the Syracuse academy. 

 Mr. J. Forman Wilkinson of Syracuse, who was at 

 this time one of Professor Eoot's pupils, has con- 

 tributed several interesting points relating to the 

 occurrence of the serpentine. In a recent letter to 

 the writer, he says, in speaking of the different 

 localities mentioned by Vanuxem and Beck, "The 

 exact place was upon the lawn now owned and occu- 

 pied by Howard G. White. . . . The specimens that 

 you have were gathered some time between 1837 and 

 1845, probably nearer the earlier period. We used 

 to go to the bed sometimes with a pick (oftener not) 

 to gather and sort out the specimens. They were 

 found in a bed of decomposed green rock, which was 

 soft, and readily gave way under the pick. This bed of 

 green disintegrated rock extended all along the side of 

 the hill from the middle of James Street, nearly to 

 the place where Howard White's house was built. The 

 specimens were, I think, all found at the north or 

 James Street end. . . . When a trench was opened 

 for water-mains opposite, and near to this deposit of 

 serpentine {about fifty feet away), the cutting was 

 through gypsu^n." The outcrop has not been accessi- 

 ble for over forty years. 



It will be readily seen that the main point of inter- 

 est connected with this rock is its mode of origin, — 

 ■whether aqueous or igneous. It is included between 

 two beds of porous limestone or dolomite. Among 

 the dozen or more specimens in the possession of the 

 writer, there are some which show angular frag- 

 ments of this limestone embedded in the serpentine. 

 la one case these are so abundant as to afford a 

 breccia with a serpentine matrix. By far the best 

 proof of the eruptive nature of the rock fi-om which 

 the serpentine has been derived is, however, afforded 

 by its microscopic structure. The hand specimens 

 agree exactlj' with the descriptions of Vanuxem and 

 Beck. There are two principal varieties, — one a com- 

 pact, dark-green rock, in which a few bronzy crys- 

 tals are seen ; and a mottled one, occasionally stained 

 with blood-red spots. A microscopical examination 

 shows that both of these rocks are most typical rep- 

 resentatives of the class known as peridotites ; the 

 former with a slightly, the latter with a very pro- 

 nounced, porphyritic structure. The original struc- 

 ture is still perfectly preserved, althouj<h most of the 

 constituents are changed to serpentine or a carbon- 

 ate. The groundmass contains, beside these two 

 minerals, magnetite, a brown mica peculiarly char- 

 acteristic of certain peridotites, green amphibole, and 

 yellowish octahedrons which may prove to be ana- 

 tase. The porphyritic crystals have the typical 

 crystal forms of olivine or enstatite, both so perfect 

 and so sharp that they could only be tbe early crys- 

 tallizations from a fluid magma. The blood-red spots 

 are seen to be due to the common staining of altered 

 olivine crystals by iron hydroxide. The more por- 

 phyritic specimens are doubtless from the edge of 

 the mass, and the coarser- grained variety from its 

 centre. 



The evidence of the eruptive origin of the Syra- 

 cuse serpentine appears, therefore, to the writer to 

 be : 1". The microscopic structure, which shows that 

 the original mineralogical composition and arrange- 

 ment of the rock were such as are only found in 

 masses of an eruptive nature ; 2". The included frag- 

 ments of the adjacent limestone ; 3°. The last remark 

 quoted from Wilkinson's letter, that fifty feet away, 

 on the strike of the deposit, only gypsum was en- 

 countered. 



There seems to be nothing in any of the published 

 descriptions of this deposit which indicates that its 

 origin was aqueous. Such an idea, expressed by both 

 Vanuxem and Hunt, is purely a malter of opinion, 

 unsupported by any facts. 



The writer hopes soon to publish in more detail the 

 results of his study of this rock. It seems to bear a 

 strong resemblance to the carboniferous peridotites 

 recently described from Kentucky by Mr. J. S. Dil- 

 ler, of the U. S geological survey, — an opinion with 

 which Mr. Diller himself wholly concurs. 



George H. Williams. 

 Baltimore, Md., March 7. 



Thought-transferrence. 



It is always a rash course to attack other people's 

 work on the strength of second-hand reports of it, 

 and doubly so when the reports have themselves 

 been those of hostile critics. This rashness I am 

 forced to impute to 'J.J.,' the writer of a paper on 

 'Some miscalled cases of thought-transferrence,' in 

 your supplement for Feb. 4, as I cannot for a moment 

 believe him capable of the deliberate stippressio veri 

 and suggestio falsi which his attempt to explain our 

 English results by ' number-habits ' woiild otherwise 

 involve. The idea that the argument for thought- 

 transferrence has depended entirely, or mainly, on 

 exj)eriments in which one person chose a number 

 at will, and another person tried to guess it, could, 

 not survive the most cursory study of the published 

 evidence. Yet that idea, picked up by ' J. J.' from 

 an article in the JMational review, is the one on which 

 his own criticism is expressly and exclusively found- 

 ed, and which every one of his readers, if unac- 

 quainted with the original evidence or some trust- 

 worthy version of it, must at this moment be hold- 

 ing. 



As a matter of fact, this type of exijeriment 

 (though, as I shall show, ' J. J.' has greatly exagger- 

 ated its defects) has hardly ever been employed by 

 lis, and its results are a negligible quantity in our 

 case. Our published records do not include a single 

 instance in which the object to be guessed was a sin- 

 gle digit chosen by the agent. Where the number 

 contains two digits, the risk of appreciable disturb- 

 ance of the results by ' number-habit ' is of course far 

 less ; and trials of this type form between a sixth 

 and a seventh part of the tabulated Creery aggre- 

 gate.' 



But their importance in the cumulative result of 

 those experiments is very much smaller than this 

 fraction would indicate ; since the success obtained 

 in them, though very remarkable, was less so than in 

 some other types. If ' J. J.' likes to omit them, one 

 and all, as ' vitiated,' he is welcome to do so ; and he 

 will, at any rate, have the satisfaction of striking a 

 certain number of noughts off the odds — estimated 

 at about a hundred million trillions to 1 — against 

 obtaining by accident the amount of success re- 



1 This aggregate consists of results where the object of 

 which the idea was to be transferred was known only to 

 some member or members of the investigating committee. 

 See the table in ' Phantasms of the living,' vol. i. p. 25, as to 

 which it should be noted, that in th-e espei'iments with sin- 

 gle digits, included under the second head of Dublin ex- 

 periments, the numbers were drawn at random out of a bag. 

 Trials with " letters of the alphabet, and names of people 

 and towns," by the way, find no place in this crucial list ; 

 bat I am curious to know whether 'J. J.' would account, 

 e.g., for the correspondences of names recorded on p. 37, 

 by 'independent similar brain-fuuctioning.' 



