MAY, 1897. MONTHLY WEATHER REVIEW. 211
the botto-m of the gauge, or the sloping fuiinel of the receiver, so far below the mouth of the gauge that drops and spatter
and hailstones can not easily bound out and be lost. I n order to catch and measure hail separate from the water,
or in order to prevent the hail from meltiug and becoming
indistinguishably mixed with the rain, some special form of
gauge is needed, such as has not yet been invented and we
commend this problem to the ingenuity of our readers. A
1.ayer of some soft substance a t the bottom of a simple cylin-
drical gauge, such as we use for catching snow, would prob-
ably prevent the loss of the hail by the rebound or the break-
ing of the hail by a violent shock, but it would not prevent the melting of the hail by the rain that usually falls with it.
As an experiment we think it would be worth while to try B
separate special hail collector to consist of a cylindrical bag,
5 or 8 inches in diameter and 2 feet long, hanging freely sus-
pended from a firm ring or hoop fastened horizontally be-
tween two posts at a few feet above the ground. The wind
will deflect such a bag from the vertical, so that hail falling into it will be apt to strike the sides and glide to the bottom
with diminished momentum without breaking; the rain that
falls will of course pass through the bag without melting
much of the hail, and, in fact, if the observer is at Iinnd, he can rescue the hail and measure it promptly before much loss
has occurred.
One df the aurious phenomena with regard to hailstones is
the fact that a t the center each stone includes a bubble of gas under-very great pressure. It irJ worth while to melt hail-
stones in a mixture of soap and water, and observe the rela-
tive diameters of the bubbles of air when inside the hail-
stoneb, and again after they have been liberated. The sudden expansion of the bubbles as they escape has been found to indi-
date that the air is imprisoned under a presmre of several
atmospheres. This could only happen in case the hailstone
is made of water that has been frozen from the outside in-
ward, thus driving ita imprisoned air to the center. Another evidence of the pressure existing within a hailstone is said to
be shown by examining the optical properties of a section, as
can easily be done by using a beam of polarized light.
IGNIS FATUUS OR JACK-0-L-N.
This title is given to flickering flames and dancing balls of
fire seen a t nighttime in marshy places. The phenomenon appears to be rare in the United States, but common in some
parts of Europe, probably owing largely to geological pecu-
liarities as affecting the nature of surface soil. The light is undoubtedly caused essentially by the slow oxidation of gases
containing some combination of phosphorus. Such gases, of
course, result from the decomposition of animal and, more
rarely, of vegetable matter: This is probably the explanation of a phenomenon recorded in the Evening News of Detroit,
April 6, as having been observed near Lee, Mich. The news-
paper account says :
Between 10 and 11 o’clock the other night a bright light was seen emerging from the river [possibly the galamazoo River in southeastern Michigan]. On first sight it waa thought to be a lantern, but further investigation proved it to be a ball of light about as large as a large hen’s egg floating through the air, about 10 feet from the ground, with whizzing sound and zigzag motion. It soon disappeared.
Although, under some circumstances, there occurs a form
pf lightning electric discharge known as “ ball lightning,” yet it is not likely that this was the case in the present in-
stance. Both the ball lightning and the ignis fatuus belong
to the rare and curious phenomena of meteorology. AI though
they have no important relation to climatology or to dynamic meteorology, yet they are always worthy of record. From
the standpoint of the electrician, ball lightning is a phenome-
hon whose nature is as yet totally uiiknowii, and a satiefac-
tory explanation thereof is greatly deeired.
UURRENT TKEATHER AND FUTURE UROPS.
An average state of weather is expected to produce an average crop and when some condition that seems abnormal
occurs, the people are full of apprehension that the crops will be greatly diminished and of inferior quality; prices go
up, speculation is rife, and the croakers have it all their own
way. But after a few weeks nature restore8 the injury that was done, and before Thanksgiving day comes around those
early fears are all dissipated by the sight of the bountiful crops. The really serious injuries to the crops almost inva-
riably occur late in the growing season, when there is no time
left to repair the damage.
Mr. J. M. Broadfield publishes several illiistrations of this
principle in a letter to Mr. George E. Hunt, Director of the
Georgia Cliniate and Crop Service, and published in the
Georgia Review for June 16, 1896. Mr. Broadfield says :
The year 1818 waa very fatal to all crops; no rain from the last of March till August; 1839, no rain from the 1st of April till 3d of July, and every farmer gave it up, that it was impossible to make but little, if anything. But the rains set in the 3d of July, and it rained every day for two weeks, and, to the astonishment of all, more cotton was made that year than an previous one. Corn took on new life, and a ver heavy crop was mde. In 1845 the drou ht set in about the laat of &arch or 1st of April, and no rain till midjle of August. Farmers rlanted corn, peas, turnips, etc., after rain set in, and made enough to rttten hogs-from the late planting. I remember we had no frost that fall till 28th of November.
April and May, 1896, were the next most remarkable de- partures from the normal weather conditions.
SEUULAR UHANGES IN CLIMATES AND CROPS.
The meteorologist appeals to his records of observations in
order to detect any change in climate, but the agriculturist naturally puts more faith in the appeal to the records of
crops and vegetation. The latter may he called a practical test of the permanency of climate, but it is also very liable
to be a deceptive one. The thermometer is a very simple
instrument compared with a plant. The records of freezing temperatures apply directly to the climate while the records
of frost-bitten plants must be interrupted by taking into
consideration the nature of the plant, its stage of develop-
ment, the moisture in the ground, the dryness and windiness
of the air. The principal uncertainty with regard to the record of a thermometer relates to our possible ignorance of its height above the ground and the extent to which i t is
shielded from radiation of heat. On the whole it must be
confessed that the imperfections of thermometric records are
quite serious and that when it comes to a question of what the climate was fifty or a hundred years ago phenology has
about as much weight as thermometry.
But any record of any climatic feature is sure to show a
wide range of extremes in the course of fifty years, and the question of a real change in climate can not be settled by
quoting a few such extremes. It has been well pointed out by Professor Bailey, in the MONTHLY WEATHER REVIEW for
September, 1896, p. 330, that phenological records have no special value to the botanist or botanical physiologist, but
their proper use is to determine average climatological con- ditions. If, for instance, we knew the average date of leafing
or blooming or ripening of any plant for the past fifty years,
and again for the preceding fifty years, the comparisons of
these averages, having proper regard to the index of annual variability, would give as clear an idea of the possible change
in cliniate as if we had correspo~iding records of the tempera-
ture, sunshine, and rainfall. It is true that the cliniate has
made the plant, and that if we knew enough about the phy- siology of plants, we might utilize meteorological records to
explain botanical peculiarities, but, practically, we can not do
this with any safety. The phenologist must be allowed to
consider his observations of plants as being a record of