February 3, 1893.] 



SCIENCE. 



65 



Recognizing from liis letter that he was writing of a case of the 

 formation of the rather rare "snow-rollers," I wrote immediately 

 requesting details, and received in reply the following letter: — 



MiLLEDGEVlLLE, O., Feb. 7, 1893. 

 Professor E. W. Claypole. 



Dear Sir: On the morning of Jan. 30, 1892, a curious phenome- 

 non was witnessed here — snow-rollers — of which I send account. 

 I found it difficult to obtain trustworthy information as regards 

 extent of area. None of the city papers spoke of that, and only 

 quoted from local county papei-s, the correspondents of which 

 furnished all that I saw regarding the occurrence. The rollers 

 may have extended over a very large area in southern Ohio. 

 They were formed in the streets of Wilmington, in Clinton County, 

 though they are not mentioned in the adjoining country. Wil- 

 mington is twenty miles from here. No one residing a mile from 

 here, in any direction, with whom I have spoken, witnessed the 

 phenomenon. 



People here are divided in opinion whether they fell or were 

 formed by rolling. Our local editors alluded to their correspon- 

 dents who spoke of the balls as bearers of " fish-stories,"' refusing 

 to believe them. 



The morning of Jan. 30, 1892. presented in this vicinity a phe- 

 nomenon of nature as striking as it is rare. The surrounding 

 clean, level fields were covered with balls of snow, varying in 

 size from three to five inches long and from one to two inches 

 wide. Wheat-fields and meadows abounded with these balls, and 

 suggested, at first sight, that a troop of school-boys had been hav- 

 ing a battle with the snow. 



Two fields, of thirty acres each, that came under my observa- 

 tion (one a new-sown virheat-field and the other a meadow) were 

 literally covered with these "snow-rollers," there being at least 

 500 on the acre. Roadsides and lots contained a few, and, what 

 is remarkable in this connection, I noticed them on housetops and 

 straw-ricks. 



On close investigation, I found the balls to be uniformly light 

 and fragile, so that to lift one and preserve its form was impossi- 

 ble. Some were oblong, some almost spherical, while others re- 

 sembled a tea-cup or small bowl. 



There were no tracks behind them, or, if these had been made, 

 the falling snow had obliterated them. 



The accompanying weather conditions were as follows: The 

 ground had been covered with snow for three weeks. A crust 

 had formed on the top, thick and firm enough in places to bear 

 up a person. This thawed a little during the afternoon of the 

 29th. The ensuing night was warm, the mercury registering 

 40' F. By ten o'clock a brisk wind was blowing, which increased 

 in velocity, and soon the snow began to fall in large, moist flakes. 

 The morning showed that about a half-inch had fallen on the 

 crust, and on this lay the balls. 



The phenomenon was reported from several places in this 

 vicinity, chiefly in Fayette County, and from Clinton County, 

 which adjoins it on the west, but nowhere did the rollers extend 

 uninterruptedly over any great area. W. S. Ford. 



In reply to a later enquiry, Mr. Ford informed me that no one 

 took a photograph of this interesting occurrence. This is much 

 to be regretted, as I am not aware that a view of any kind is in 

 existence showing these snow-rollers. The chance of obtaining a 

 unique negative was lost. 



I cannot say if the phenomenon here described is really so rare 

 as the scarcity of published accounts would lead us to infer. Per- 

 haps the publication of this note may lead to clearer knowledge 

 on this point. 



Not having access to the earlier scientific literature on the sub- 

 ject, I am able to quote only a few instances of snow-rollers. 

 Several years ago there was a short correspondence in the columns 

 ■of Nature, from which I condense the following statements. 



In the issue of March 29, 1883, Mr. G. J. Symons wrote that he 

 believed that the first recorded account of the phenomenon 

 appeared in an early number of the Philosophical Transactions, 

 from the pen of Dr. Clouslon of Sandwich Manse, in the Orkney 

 Islands. 



Mr. Symons adds that he has heard of but one case in England. 

 This was reported by Admiral W. F. Grey in the Meteorological 

 Magazine for May. 1876. It occurred on his lawn in the south-east 

 of England, and the balls or rollers varied from the size of an 

 ordinary snow-ball to that of a cubic foot, and each one left a de- 

 cided track to the leeward. In this case they were seen to form 

 in the evening. 



The correspondence above quoted was called out by a letter 

 from Mr. S. Hart of Trinity College, Hartford, Conn., to the edi- 

 tor of Nature, mentioning that on Tuesday evening, Feb. 22, 1883, 

 a light, damp snow fell on the crust that had formed over the 

 earlier snowfall, and that a brisk wind sprang up after midnight. 

 Next morning on the college campus and in the adjoining park 

 and empty lots numbers of natural snow-balls, spherical and 

 cylindrical, from twelve to eighteen inches in diameter, and hol- 

 low at both ends, were strewn over the ground. Behind them 

 were visible to leeward tracks 25 or 30 feet long in the new-fallen 

 snow. The rollers were so light and fragile that handling was 

 impossible. A few of them could be traced 60 feet and some had 

 even been rolled up hill. They were especially fine on the frozen 

 Connecticut River. 



The same writer also reports a similar occurrence in New Jer- 

 sey in 1808, when the rollers were seen to form in the day-time, 

 and extended over at least 400 square miles of country. 



In a subsequent letter, printed in the number of Nature for 

 March 6, 1884, Mr. Hart returns to the subject and reports a repe- 

 tition of the phenomenon in Oneida and Herkimer Counties, in 

 New York, on Tuesday, Jan. 22, 1884, adding that the rollers 

 were of the same size as in the former case, but were in some in- 

 stances firm enough to be picked up and handled without crum- 

 bling. This is, so far as I know, the only occurrence in which 

 this has been reported possible. 



The wide experience of Mr. Symons is sufficient ground for as- 

 suming that the formation of snow-rollers is not frequent in Great 

 Britain, and the scarcity of records here leads to a similar con- 

 clusion for this country. Further reports on the subject are de- 

 sirable. 



LETTERS TO THE EDITOR. 



t". Correspondents are requested to be as brief as possible. The writer's name 

 is in all cases required as proof of good faith. 



On request in advance, one hundred copies of the number containing his 

 communication will he furnished free to any correspondent. 



The editor will be glad to publish any queries consonant with the character 

 of the journal. 



Some Detailed Evidence of an Ice-Age Man in Eastern 

 America. 



Mk. Holmes's statement in his communication to Science for 

 Jan. 20, that "If there was, as is claimed, an ice-age man or at 

 any rate a palaeolithic man in eastern America, the evidence so 

 far collected in support of these propositions is so unsatisfactory 

 and in such a state of utter chaos that the investigation must 

 practically begin anew," should not be allowed to go unchallenged. 

 I will content myself, however, by giving the details in a single 

 case, namely, those concerning the implement which was found 

 in 1889 by Mr. W. C. Mills at Newcomerstown, Ohio, and which 

 is now in the collection of the Western Reserve Historical Society 

 in Cleveland. 



Though the discovery was made in October, 1889, it was not 

 brought to public notice until the next spring, when I chanced 

 to meet Mr. Mills and learned about it. He then forwarded it to 

 me, when its exact resemblance in form and finishing to an im- 

 plement which I have in my own collection, that was obtained 

 by Dr. Evans of London at Amiens, France, greatly impressed 

 me. I forwarded it immediately to Professor H. W. Haynes of 

 Boston, whose expert judgment is second to that of no other per- 

 son in America, or indeed of the world. Professor Haynes ex- 

 hibited it at the meeting of the Boston Society of Natural History 

 on May 7, 1890, and his account was published in the Proceedings 

 of that evening. In conclusion, after having enumerated its 

 distinctive characteristics, he said, " I desire to express most em- 

 phatically my belief in the genuineness and age of this Newcom- 

 erstown implement, as well as to call attention to the close re- 



