86 SCIENCE. 
The full significance of the result of these in- 
oculations was not realized at the time since it 
was not then known that Sphzropsis occurred 
on these blighted areas. In the spring of the 
present year, however, a Sphxropsis was found 
to be comparatively abundant on the diseased 
bark of pear trees in the station orchards. 
Since that time a large number of pear trees 
from various localities affected with body 
blight have been examined and in nearly every 
instance this fungus was found to be more or 
less abundant. One case particularly worthy 
of notice was that of a comparatively young 
orchard that was severely attacked by body 
blight and the fruit of a Sphzropsis was so 
abundant that the conclusion was irresistable 
that this fungus must be the cause of the disease. 
Numerous inoculations made this spring with 
cultures of the Sphzropsis in large trees and in 
nursery stock clearly show that this fungus 
may produce body blight of pear trees. 
Other species of fungi closely associated with 
the Spheropsis frequently occur on trees at- 
tacked by body blight, Macrophoma malorum 
(Berk.) Berl. et Vogl. being specially abundant. 
The studies have not yet progressed far enough 
to determine what part these other fungi play 
in producing the diseased condition. Bacteria 
may also be concerned in this trouble, but of 
this we have as yet no proof. 
W. PADDOCK. 
GENEVA, N. Y. 
FORMATION OF CUMULUS CLOUDS OVER A 
FIRE. 
In Scrence of January 8, 1897, Mr. R. DeC. 
Ward describes the formation of cumulus 
clouds over a fire in Cambridge. Last Friday 
(June 30th) another phenomenon of this kind 
was observed at Blue Hill and from Winthrop 
and approximate measurements of the height ob- 
tained. 
The fire was in South Boston and consumed 
the buildings of the Bay State Iron Works. The 
smoke cloud was not of unusual size, but rose 
vertically to a considerable height (800 to 1,000 
metres), encountering at this height a north- 
westerly wind, which swept it nearly horizon- 
tally over the harbor. The fire began before 8 
p. m., and the smoke reached its greatest 
height about 8:05 p. m. At 8:03 p. m. a 
[N. S. Von. X. No. 238. 
small white cloud began to form at the apex of 
the smoke, which at this time was apparently 
nearly over Long Island, in Boston Harbor. The 
cloud increased rapidly in height, assuming the 
form of a true cumulus and reaching its great- 
est size at 8:05 p. m. The accompanying 
sketch shows roughly the appearance of the 
smoke and the cumulus at that time. The sky 
BosTon. iz 
CUMULUS FOR/TING OVER FIRE, JUNE 30,1899. 
was nearly clear, no other low clouds being in 
the vicinity of the smoke. The cumulus cloud 
is shown at (A) and apparently was about 3° in 
height and length, the highest or thickest end 
being toward the north. Between 8:05 and 
8:07 p. m. another smaller cloud formed at the 
edge of a rift in the smoke considerably lower 
than the one just described. Its position is 
shown at (B). Mr. A. E. Sweetland, of this 
observatory, who at this time was in Winthrop, 
about 5 miles (8 kilometres) northeast of the 
fire, estimated the altitude of the highest cumu- 
lus to be 15°, while a measurement made with 
the nephoscope at Blue Hill, about 10 miles 
(16 kilometres) south of the fire, gave 10° as its 
altitude as seen from the observatory. These 
measurements show that the vertical height of 
the top of the cloud at A was at least 2,500 
metres, while that of the cloud at B was about 
2,000 metres, above sea level. 
The smoke began to diminish in quantity at 
8:07 p. m. and separated from the clouds, which 
became flatter and more elongated. At 8:11 
the clouds were separated from the smoke by a 
space several degrees wide, and after this time 
they slowly evaporated. 
S. P. FERGUSSON. 
BLUE HILL OBSERVATORY, July 1, 1899. 
A REPLY TO MR. MARLATT’S ARTICLE ON SOURCES 
OF ERROR IN RECENT WORK ON COCCID.* 
WHEN I lived in Colorado, some years ago, I 
remember hearing it said that a man who had 
* SCIENCE, June 16, 1899, pp. 835-837. 
