Mar. 1902.] BOTANICAL SOCIETY OF EDINBURGH 255 



The storm burst with a deafening peal of thunder 

 directly overhead, and this was almost immediately 

 followed by the hail. In a few moments the air was 

 full of twigs and leaves which had been cut off the trees. 

 Many of the hailstones which worked this havoc were 

 one and a half inch in their longest diameter, while on 

 an average they measured fully one inch. I have been 

 told, on good authority, that some measured one and three- 

 quarter inch in diameter. The structure of these hail- 

 stones was curious. They were all more or less flat in 

 shape, the shorter diameter being about half the length 

 of the longer. One surface was concave and the other 

 was convex. Most of them had an onion-like structure, 

 being composed of successive layers of clear and of 

 opaque ice. 



The thunderstorm was quite local, embracing a width 

 of two miles and a length of about ten miles. To give 

 some idea of the force of the storm, I may say that 

 over one thousand square feet of plate glass (21-ounce), 

 representing a quarter of the whole glass area in the 

 greenhouses here, were broken. The opposite side of the 

 valley, however, seemed to catch the force of the storm 

 even more fully. At one place, roofing glass nearly half 

 an inch thick was broken ; and I noticed a field of early 

 potatoes, which were nearly ready for raising, completely 

 stripped of their leaves, and with only a few bare 

 haulms left standing. The main roads, which are but 

 slightly sheltered by trees, were strewn so thickly with 

 twigs and leaves that these latter had to be swept up 

 into heaps about four feet in height, and at intervals of 

 about seventy-five yards. Many twigs were left hanging 

 on the trees, which, after a few days, presented quite a 

 withered appearance. 



The storm, coming as it did in the middle of June, 

 caught the trees when in their first full leaf. Thus 

 numbers of the still tender twigs were cut clean off, and, 

 owing to the loss of leaves, and the lessening of the 

 feeding area, the year's growth must have been corre- 

 spondingly reduced. 



The wounds caused on the stems of trees were observed 

 only on young bark which was still smooth, and no 



