SCIENCE. 



567 



tive as distributors of heat. The temperature would 

 hence become approximately proportional to the solar 

 accession, which has been computed by Much for the 

 various latitudes, and may be roughly reduced to ther- 

 mometrical degrees by means of an easily determined 

 constant. Moreover the presence of the ice would 

 greatly facilitate both radiation and direct reflection of 

 solar energy. The general diminution of temperature 

 produced in this manner is calculated for each latitude ; 

 and from a comparison of these figures with actual tem- 

 peratures, as recorded by Dove, the temperature of the 

 whole hemisphere when the ice-sheet extends to any lati- 

 tude is also computed. From the several figures ob- 

 tained it appears that if the globe were encrusted with 

 ice, the crust would probably (and indeed almost cer- 

 tainly) never be melted unless by proper terrestrial heat ; 

 while the temperature in polar regions, as well as over 

 much of the ice-covered hemisphere, would sink so low 

 as to practically eliminate all aqueous vapor and effectu- 

 ally prevent the further accumulation of ice. The annual 

 variations in solar intensity would not materially affect 

 the values obtained. 



Since the results reached in the manner indicated em- 

 body values widely different from those of existing tem- 

 peratures, and are hence a priori improbable, a detailed 

 investigation of certain meteorological phenomena is un- 

 dertaken in, order to verify these results. The obseived 

 and computed temperatures of the northern hemisphere 

 are first compared, and are found to indicate that the 

 temperature-equalizing agencies are 1.5 times as effective 

 in summer as in winter. The effect of atmospheric dry- 

 ness in diminishing the effectiveness of these agencies is 

 then found to be still greater. The values developed in 

 the investigation of this subject demonstrate that the 

 climatal perturbations previously pointed out as the 

 necessary result of the considerable extension of a polar 

 ice-sheet do not differ in kind, but only in degree, from 

 those whose constant occurrence is a matter of author- 

 itative record ; and analogy with observed phenomena 

 moreover indicates that the calculated extent of these 

 vicissitudes is in perfect harmony with the magnitude of 

 the formulated course. 



The figures obtained incidentally demonstrate the 

 existence of an empirical meteorological law, which may 

 be stated as follows : Any increase in thermometri- 

 cal range is accompanied by a diniinutio7i in mean 

 temperature. Since the law strongly corroborates the 

 results .eached by the second line of investigation, it 

 is quite fully considered, especially in its application to 

 the present condition of the two hemispheres. That 

 hemisphere whose winters occur in aphelion experiences 

 a greater variation in solar accession and consequently 

 in temperature than the opposite one, and hence, accord- 

 ing to the law, ought to have a lower mean annual 

 temperature. The southern hemisphere is so situated 

 at present ; and accordingly, notwithstanding more 

 favorable geographical and other conditions, its tem- 

 perature is lower than that of the northern hemisphere. 

 The bearing of the law on Croll's celebrated theory of 

 secular variations in terrestrial climate is manifest. 



Since it is developed in both lines of investigation 

 that the accumulation of glacier ice is dependent upon, 

 and in a general way proportional with, precipitation, 

 the maximum accumulation at any latitude may be 

 roughly computed. The final determination is as follows : 



Latitude 40 18,594 feet. 



V* 5° 9.777 " 



" 60 5,728 " 



" 70 2,800 " 



" 80 1.799 " 



" 90 1,440 " 



It may accordingly be concluded that a sufficient 

 accumulation of polar ice to displace seriously the 

 ■earth's center of gravity, or to influence the motion of 



middle-latitude glaciers, never can have taken place. 



The nature and course of ice motion are discussed 

 at some length; and the phenomenon is shown to be 

 analogous to those exhibited by all classes of sub- 

 stances, though generally in a less striking degree. 

 The "viscous theory" of Forbes is adopted with some 

 modifications ; and the principal objections thereto are 

 considered. It is also pointed out that ice-streams are 

 necessarily in tension, and hence that the central mass 

 of an ice-field can exercise an influence on the motion 

 of its peripheral portions. The assumption of a vast 

 polar ice-cap to explain the motion of the quaternary 

 glaciers accordingly appears to be not only unneces- 

 sary but incompetent. 



GLUCOSE. 

 By Albert E. Ebert, Peoria, III. 



The process of making glucose, or grape sugar, is as 

 follows: corn, after being shelled, is placed in large tubs 

 and soaked in hot water from a day and a halt to five 

 days, or even longer, .the time depending on the hardness 

 of the grain. If termentation is not wished, the water is 

 changed when the substance begins to sour. It is then 

 ground, while wet, with ordinary burr stones, and with 

 a stream of water running into the hopper with the corn. 

 The meal, or "chop," is then run on vibrating sieves, 

 made of fine silk bolting cloth, also fed with streams of 

 water. By this treatment the starch, which washes 

 through the sieves, is separated from the gluten and cel- 

 lular matter, which waste portions go over the tail of the 

 sieves, and after passing through rollers which squeeze 

 the mass, and return the water to the sieves, is sold for 

 feed. The portion which went through the sieves is run 

 into tanks and settled, the water drawn off and the sedi- 

 ment again mixed with clean water and treated with 

 alkali, about one pound of caustic soda, (more or less, 

 according to the hardness of the water), being used for 

 each bushel of corn. This treatment separates any traces 

 of gluten from the starch, which is then run into metal- 

 lined troughs or gutters about eight inches deep, from 

 fifteen to thirty-six inches wide, and usually from one 

 hundred to one hundred and fifty feet long. These are 

 inclined slightly, and the water runs off at the lower end, 

 leaving the starch as a sediment at the bottom. In some 

 factories this starch mixture goes direct from the sieves 

 to the gutters or " tables," as they are usually called. It 

 is left to dry somewhat in the tables, and is then shoveled 

 out. At this stage of the process it is known as " green 

 starch." It is quite solid, but moist, containing about 

 fifty per cent, of water. 



Up to this point the process is the same as starch- mak- 

 ing. Starchmakers take the green starch and wash it, 

 some several times, by mixing it with clean water and 

 allowing it to settle, then drawing off the water, and re- 

 peating the process. It is then sometimes bleached by 

 chloride of lime or sulphurous acid, and after washing, 

 settled, made into blocks about eight inches square, when 

 it is dried in a kiln. For the finer grades, about half an 

 inch of each side of the cake is shaved off when partially 

 dry, the rest of the cake being wrapped in paper and put 

 back into the kiln until it forms into little sticks or pipes. 



For glucose, however, the green starch is made quite 

 thin with water, and run into converters, usually after 

 several additional washings. The converters are large 

 wood tanks or tubs, where it is treated with acids, sul- 

 phuric being usually used, although muriatic, nitric, or 

 even oxalic may be substituted. Sulphuric is preferred, 

 as it is cheap and easily gotten rid of in an after stage of 

 the process, when it has performed its work. The acid 

 does not combine with the starch, but merely exerts a 

 catalytic action ; therefore the necessity of providing for 

 its removal. While under the acid treatment the con- 

 tents of the converters are heated to the boiling point by 



