558 POPULAR SCIENCE MONTHLY 



sedimentary deposits consisting of clay, sand, sand and gravel, coarse 

 gravel and finally large pebbles. 



Over the outer portion of the sandy bottom, also for great dis- 

 tances beyond, as well as up and down the river wherever extensive 

 silting has developed the formation of muddy bottoms, there is an 

 abundant growth of eel grass (Zostera marina) which, together with 

 other debris of a similar nature, is continually washed upon the beach, 

 broken up by the combined action of the waves and sand and gradually 

 buried in the latter, so that each year the deposit of organic matter 

 is increased by definite though rather slight increments. 



From these data it will be observed that some special significance 

 attaches to the fact that the fire, on two separate occasions, was strictly 

 confined to the beach, and that it did not in any way extend over the 

 limiting areas of rock. 



On the evening of Friday, September 1, 1905, the guests in the 

 hotel, the piazza of which may be seen on the extreme right of the 

 photograph, were startled by the appearance of flames rising from the 

 beach and also from the surface of the water. The tide was about one 

 hour lower than shown in the photograph, so that a very considerable 

 portion of the sand was uncovered. The conflagration occurred be- 

 tween seven and eight o'clock in the evening and lasted for upwards 

 of forty-five minutes. It was accompanied by a loud and continuous 

 crackling noise, which could be distinctly heard one hundred yards 

 distant, due to the rapidly recurring explosion of bubbles of gas as 

 they came to the surface of the sand or water. At the, same time there 

 was a very strong liberation of sulphurous acid gas, which penetrated 

 the hotel, drove the proprietor and his staff from the office and filled 

 the other rooms to such an extent as to cause great inconvenience to 

 the guests. So great a heat was developed that the sand could not be 

 held in the hands, while sand placed in a tumbler with water and then 

 stirred, liberated bubbles of gas which ignited upon coming in contact 

 with the air. On this occasion the fire developed over that portion of 

 the sand which had been exposed by the falling tide, and it also ex- 

 tended out over the water for a distance of thirty or forty feet. 



On the evening of Wednesday, October 4, 1905, as reported by a 

 reliable observer, the phenomenon was repeated with identical features, 

 except that instead of occupying the entire area between the rock for- 

 mation on each side, it was restricted to the area where the two boats 

 are lying. It therefore occupied probably less than one fourth the 

 area of the first conflagration. 



It is difficult to estimate the height of the flames on these two occa- 

 sions, since the conditions under which the fire occurred would tend 

 to give an exaggerated value. It is probable that in general the flames 

 were not more than three or four inches in height, and this would 

 be a reasonable estimate when arising from small bubbles of gas. 

 But, as stated in the original account, the flames attained a maximum 



