December 4, 1!I03.] 



SCIENCE. 



r.n 



13. The outwash deposits of the two Wis- 

 consin stages are likewise relatively thin, being 

 generally underlain at slight depths by one 

 or the other of the older formations described. 



14. Three groups of artesian wells have been 

 recognized: (1) the deep Cretaceous wells, 

 (2) wells in the Jameco gravels, and (3) shal- 

 low wells in the Pleistocene deposits of the 

 north shore. 



15. The deep Cretaceous wells are found 

 both on the north and south shores. In both 

 localities the water horizon has a regular 

 southeastward dip. The source of the water 

 in the south shore wells, except in the Barren 

 Island well, is probably in the highlands of 

 the island itself where the beds rise to the 

 surface. In the north shore wells and in the 

 Barren Island well, which perhaps obtains its 

 supply from the same water-bearing horizon, 

 the original source of the water is not yet 

 established, although the problem is under 

 investigation. 



16. The Jameco wells obtain their supply 

 from the glacial gravels (Kansan) occupying 

 the deep and well-defined channel extending 

 beneath Jamaica and Jamaica Bay and under- 

 lying the thick clays of the succeeding (Yar- 

 mouth) interglacial stage. The supply is de- 

 rived from the ground water entering the 

 gravels under the landward edge of the over- 

 lying clay. Because of the coarseness of the 

 gravel the water is given up freely. 



17. The shallow north shore artesian wells 

 are generally restricted to the upper halves 

 of the deep reentrant bays, generally at the 

 base of steep slopes. The source of supply 

 is from the ground water of the glacial grav- 

 els and sands, the flow taking place by virtue 

 of the freer passage afforded by the wells than 

 by the gravels. 



18. The great thickness of the sandy layers 

 of the Cretaceous under the higher portions 

 of the island, the extent to which the Cre- 

 taceous water-bearing sands have already been 

 developed and the probability that a num- 

 ber of water horizons have been previously 

 overlooked because of search for a coarse 

 gravel like the Jameco, makes the recom- 

 mendation made by Professor C. S. Slichter 



regarding the advisability of sinking deep 

 wells with casing perforated at each water- 

 bearing horizon particularly pertinent. 



M. L. FULLEK, 



A. C. Ve.\tch. 



CURRENT N0TE8 ON METEOROLOOY. 

 BLOOD COUNTS AT HIGH ALTITUDES. 



Nos. 8 and 9 of Vol. HI., Bulletin of the 

 Hadley Climatological Laboratory of the Uni- 

 versity of New Mexico, deal with ' Cold as a 

 Causal Factor in the Blood Changes due to 

 High Altitude,' and with ' Further Observa- 

 tions on Increased Blood Counts due to High 

 Altitude.' The first paper is by John Wein- 

 zirl, M.S., and the second, by the same author 

 with the cooperation of C. E. Magnusson, 

 Ph.D., is a study the prosecution of which was 

 aided by a grant from the Elizabeth Thomson 

 Fund. The fact of an increased number of 

 red blood corpuscles at high altitudes is well 

 known, and has been investigated by Bert, 

 Egli-Sinclair, Viault, Miintz, Egger and 

 others. By means of blood counts in the 

 cases of human beings and of rabbits, the 

 authors of these papers come to the conclu- 

 sions that cold is an important, though not 

 the only, factor in producing blood changes 

 at high altitudes, and that the increase in the 

 number of red corpuscles due to altitude is 

 temporary (as, it should be noted, has already 

 been showoi by several writers), this temporary 

 increase being very largely due to the change 

 in the temperature and not to the diminished 

 pressure. 



WEST INDIA HURRICANES. 



Ox the ' Pilot Chart of the North Atlantic 

 Ocean ' for September last there is a brief 

 but well-arranged summary of the most im- 

 portant facts regarding West India hurri- 

 canes, prepared by James Page, of the United 

 States Hydrographic Office (reprinted from 

 H. O. Publication, No. 86). Of .56 hurri- 

 canes recorded by the Hydrographic Office 

 between 1890 and 1900, 41 occurred in Sep- 

 tember and October. Instead of the old 

 ' Eight Point Rule,' which is now known to 

 hold only for the central portion of the storm, 

 seamen are at present instructed tliat ' six 



