380 Tin' Aiini'ictUi (j'rtilixii.sf. J).-ci'iiil>ii, IHJil 
jVIost of the very pU'iitiful Ixuildcrs usually pi-cscnt in the hills 
and ri(l*;os of the Icnniiml uioiaiiu's wore douhtlcss Itrouoht as 
the higlu'st part of the cuiilacial drift. 
K'niics. Osdis, iiihI Valhii l>r!ft. That a lar^o amount of drift 
was carried along in the ice-sheet and Ix'canie exposed on its sur- 
face during' its linai nieltin<>\ seems to me clearly proxcd. not 
oidy by the thickness of en>;lacial till according to the criteria he- 
fore noted, hut also l)y the various deposits of assorted or modi- 
tied drift. Perhaps the most interesting class of these deposits is 
that which consists of i)rolonged ridges of irregularly hedded gravel 
and sand, often extending in a series many miles, sometimes 20. 
.')(! or ev(Mi KM) miles or more in length. These ridges usually 
have stee[) sides antl a nairow arched crest of varial)le hight. 
As.sociated Avith them. an<l especially with tlu> terminal and mar- 
ginal moraines of the ice-sheet, are mounds, iiillocks. and short 
ridges likewise composed of gravel and sand iiaviug a confused 
stratification, often somewhat anticlinal in conformity with the 
slopes of the surface. Uoth the veiy long gravel ridges or series 
t)f ridges, and the very short ridges, hillocks, and knolls, were 
formerly cla.ssed together, and were called kames. eskers, or 
osars. hut :i usefid discrimination has Iteeu })rop()sed liy McGee 
and (Miand)erlin, in accordance with which the term Ldiius is now 
restricted to the gravel hillocks, knolls, and ridges of sligiit ex- 
tent, while the long ridges are named osm-s or rs/,-trs/'^ 
Precisely the same explanation of the mode of formation of the 
osars was reached independently fifteen years ago by ])r. X. 0. 
Hoist in Swedent and Ity the present wiiter in New Hampshire, t 
Four years earlier, as I afterward leariu'd. nearly the same view 
had been first published by Prof. N. 11. M'inchell, in 31innesota.i| 
*W J McGee, in the Report of the International Geological Congre.ss, 
second session, Bouloune, 1881, ]>. 021. T. C. Clianiberlin, in the Third 
Annual Report of tlii' 1'. S. (;eoJ. Survev for 18Sl-'S2, p. •i'J'.l; and Am. 
.lour, of Science, 111, vol. \xvii, pj). :n.S-:UIO, May, 1S81. The article last 
cited presents many hihlioiirapliic references, and shows that tin- term 
osar (pi. osars I, in tins Anglicized form, has long been in common use l)y 
Jackson, Ilitcluock, Desor, Mnrehison, and other authors. 
t"()m de glaciala rullstensasarne,"' (ieologiska Foreningens i Stcrk- 
holni Fcu'handlingHr, vol. iii, 18T(i, pp. 1(7-112. Tins paper is reviewed by 
Dr. .losua Lindald in the American Naturalist, vol. xxli, jip. ."(H!)-.")!*! and 
',M-1V\, July and Aug., t8S8. 
:5:"On tlie Origin ol Ivanies or Kskers in New Ilanishire," Proe. A. A. A. 
S., voL XXV. 187i), pp. 21<i-22r). Geology of N. H., vol. iii, 1.S78, chapter i. 
Proceedings, IJoston Society of Natural nistory,vol. xxv, 1801, pp. 228-242. 
jGeol. and Nat. lli.st. Sin"vey of ^Miun., I'^'irst Annual Report for 1872, j). 
62. etc.; Second .\n. Mc]^. for 1^7:). p. 1!t4. Proe. A. .\.A. S.. vol. xxi, i'or 
1872, p. Vi'). 
