190 RECOKD OF SCIENCE FOR 1886. 



tributious are made to this subject. lu au address before the Ameri- 

 can Association in 1880, Chamberliu* summarized the results attained 

 up to that time, especially those relating to drift i)henomena. The 

 following is a very brief abstract of his conclusions : Three phases 

 are recognized in the course of the uudulatory drift border: (1) a 

 thickened edge or terminal moraine ; (2) a thin margin ; and (3) an 

 attenuated border of scattered pebbles. The morainic border prevails 

 from the Atlantic to the Ohio, while attenuated borders reach thence 

 to the Eocky Mountains. Attenuated borders delimit an earlier ice- 

 incursion, and the morainic border a later one. The interval be- 

 tween the two principal glacial epochs is represented by changes 

 of orographic attitude and drainage ; by different degrees of erosion, 

 decomposition, and ferrugination ; and by vegetal accumulations and 

 lacustrine oscillations. The chief iuterglacial epoch was marked by 

 great erosion. The drifts are grouped into the earlier, embracing two 

 or more subdivisions, and the later, embracing several subordinate 

 phases; also a third series, embracing the Great Basin deposits of 

 aqueous origin. Of bowldery clays at least three genetic classes are 

 recognized: (1) the subglacial ; (2) the englacial or snperglacial ; (3) 

 the subaqueous; and perhaps (4] tills ridged transversely by the thrust 

 of the margin of the ice. Of moraines, terminal, lateral, medial, and 



interlobate are described; and of forms e^ ouud moraine the following 



are enumerated: (1) till tumuli; (2) mammillary and lenticular hills; (3) 

 elongated parallel ridges; (4) drift billows; (5) crag and tail; (6) pre- 

 crag and combings; and (7) veneered hills; the first three being grouped 

 under the term " drumlins." Of the assorted drifts, two classes com- 

 monly embraced are excluded: (I) the "orange sands," commonly 

 regarded as Champlain de[X)sits, because there is great uncertainty in 

 regard to their age, and good reason to believe that they are not Cham- 

 I)laiu; (2) drifts reworked by uon glacial agencies. Omitting these, two 

 classes there are recognized; (1) those that gathered immediately within 

 and beneath the ice body itself, or against its margin, and (2) those 

 borne to distances beyond its limit by glacial drainage. Of the first, 

 there are the products (a) of snperglacial streams; (6) of moulins ; (c) of 

 subglacial streams; {d) of streams in ice-canons; and (e) debouchure 

 deposits at the glacial margin. The importance of the distinction be- 

 tween kames and osars is urged. Of valley drift the intermediate 

 phases are pass^ over, and attention is directed to two extreme 

 phases: {a) the moraine-headed valley trains and (b) the loess, the 

 former the deposit of vigorous glacial floods, the latter inferred to be 

 the product of slack drainage. Attention is directed to the ice-blocked 

 ancient lakes, especially of the greater basins, to the overflow phenom- 

 ena, and to the difference between ancient and existing water levels. 

 Finally, the current interpretations of glacial phenomena, and specula- 



Am. Assoc, Proc, vol. 35, pp. 195-211; Science, vol. 8, pp. 156-159. 



