MCKENZIE: STORM DAMAGE IN MASSACHUSETTS 
31 
being the greatest rainfall during any month since 1836 when records 
were first taken in Amherst. Barometer readings, when reduced to sea 
level, were fairly high from September 17 to noon on September 21, drop¬ 
ping only from 30.15 inches to 29.70 inches. At four o’clock on the 21st 
the barometer had dropped to 28.72 inches and at five o’clock the low of 
28.41 inches was reached. The wind velocity rose from 20 miles per 
hour at three o’clock to 80 miles per hour at 5:17. The wind blew 
mostly from the north during the storm on September 17, 18, 19, and 20. 
On the 21st at noon the wind shifted to the southeast and remained in 
this direction during the hurricane. 
When the high wind struck Central Massachusetts it found the stage 
well set for an all time record of tree destruction. The thoroughly soaked 
ground freely gave up the root systems supposedly anchored tenaciously 
by functional and physical laws. Rather close observation of several 
large trees preceding and during the process of uprooting revealed no 
thundering crash but rather so relatively slow and measured a fall that 
huge trees frequently fell with almost no audible indication and in some 
cases leaned over to rest against structures with a minimum of damage to 
buildings. 
Poorly developed roots, often the result of severe pruning during road 
construction, poor site, or inadequate water or food, contributed to 
materially weaken the mechanical support of trees. However, inconsis¬ 
tencies with the theory that inadequate anchorage alone determined 
victims of the gale wind are too numerous to conclude that a good root 
system insured tree survival. Likewise the existence today of many 
avenues of trees in the path of the hurricane which were notoriously 
weakened at some time in the past by root cutting and mutilation, is 
evidence that a combination of factors was active in the destruction and 
survival of trees during the hurricane. 
Further analysis of the meteorological statistics has value in surveying 
shade tree damage, at least in a negative way since some observations of 
fallen trees are not to be explained by the observatory records. The 
weather report indicates that the wind was from the southeast during 
the entire period of the hurricane. This being the case why should trees 
be blown over in almost every conceivable direction? It is known that 
the velocity of the wind varied considerably during the storm and it is 
entirely possible that the hurricane encompassed miniature tornadoes 
which broke loose in limited areas battering down whatever came within 
their paths until force was exhausted by friction. If such a condition 
