March 16, IBOO.] 



SCIENCK 



429 



of the power. The first number in the series 

 of first differences is always 1 ; the first in that 

 of the second diflTerences the difference be- 

 tween 2" — 1, and 1 or 2" — 2 ; while the third 

 series is evidently headed by 3" — 3 ■ 2" + 3, the 

 fourth by 4 " — 4 • 3" + 6 • 2» — 4, and the q'" by 



g(g-l)(g-3) 



(2-3r + 



The derivation of these numbers for the nth 

 power from similar numbers for the n — 1th 

 power is very simple, since 



g(g-l) 



, 2(9-1) 



-2)" = 



■2(4-l)"-'- 



(g-1)"- 



(2_2)«-i- 

 ■{q-l)(q-2Y 



The table below shows the succession of lead- 

 ing differences for each power as far as the fifth ; 

 figures in the column to the left denoting 

 powei-s, and each number in the body of the 

 table being the sum of that immediately above 

 it and that immediately to the left of the latter, 

 multiplied by the factor at the head of its 

 column. The calculation of a table of fourth 

 powers is also indicated to the right ; the num- 

 bers at the top being taken from the preceding 

 table, and each of the rest being the sum of 

 that immediately above it and that immediately 

 to the right of the latter. The number 24, in 

 the last column, is a constant additive to the 

 column preceding. The successive fourth 

 powers appear in the left-hand column of the 

 calculation. 



II 2 3 4 5 



A paper by Mr. J. R. Eastman on the ' Treat- 

 ment of Reflection Observations at Greenwich 

 Observatory,' announced for the evening, was 

 not read on account of the unavoidable absence 

 of the author. E. D. Preston, 



Secretary. 



GEOLOGICAL SOCIETY OF WASHINGTON. 



The 98th regular meeting was held at the 

 Cosmos Club, February 28, 1900. 



Under informal communications Mr. G. K, 

 Gilbert called attention to the peculiar level 

 character of the ledges of rock crossing the bed 

 of the Potomac, just above Harpers Ferry, and 

 also in the bed of the Columbia river, near the 

 mouth of the Umatilla. It was suggested that 

 subaerial disintegration is ettective in reducing, 

 to approximately the water level, those por- 

 tions of the rocky bed which are not ordinarily 

 covered with water. 



On the regular program Mr. W J McGee 

 presented a paper on ' The Gulf of California 

 as an Evidence of Marine Erosion.' It was 

 shown that the powerful tides of the Gulf, aided 

 by frequent gales, are the cause of vigorous 

 marine erosion where the tidal currents are 

 constricted by the islands Tiburon, Esteban, 

 and San Lorenzo. The erosion results in sub- 

 marine terraces, up to a mile in width, covered 

 with shallow water, and backed by precipitous 

 coastal cliffs. At the outer edges of these 

 terraces there is a rapid descent into deep 

 water. 



A discussion on ' The Conditions of Forma- 

 tion of Conglomerates, and Criteria for dis- 

 tinguishing between Lacustrine and Fluviatile 

 Beds,' was introduced by a paper from Pro- 

 fessor W. M. Davis, briefly summarizing the 

 criteria available for discriminating the two 

 classes of deposits, and suggesting that the 

 term continental proposed by Penck, should be 

 used in those cases where it is not possible to 

 determine whether a given deposit is lacustrine 

 or fluviatile. 



Mr. G. K. Gilbert followed with a short 

 analysis of the conditions governing the forma- 

 tion of conglomerates. The dominant agencies 

 are littoral and fluviatile. Hence the presence 

 of conglomerates, in the absence of contrary 

 evidence, indicates stream or shore action. 

 The formations of Lake Bonneville and the 

 superficial deposits of the Great Plains were 

 determined as lacustrine and fluviatile respect- 

 ively, not from the internal evidence of the 

 deposits, but on physiographic grounds. 



Mr. S. F. Emmons, referring especially to 

 the regions covered by the Fortieth Parallel 



