xt73n58ch60s https://exploreuk.uky.edu/dips/xt73n58ch60s/data/mets.xml   Kentucky Agricultural Experiment Station. 1964 journals 139 English Lexington : Agricultural Experiment Station, University of Kentucky Contact the Special Collections Research Center for information regarding rights and use of this collection. Kentucky Agricultural Experiment Station Progress report (Kentucky Agricultural Experiment Station) n.139 text Progress report (Kentucky Agricultural Experiment Station) n.139 1964 2014 true xt73n58ch60s section xt73n58ch60s LA I E S PRI N G        —   
g E A   Y   I      =rK   
· K EN T u c KY  
By Doyle Cook · Agricultural Meterologist · Weather Bureau Airport Station
Louisville
Progress Report I39 Filing code 24
UNIVERSITY OF KENTUCKY AGRICULTURAL EXPERIMENT STATION
Lexington

 PREFACE ‘
This publication is intended as a handbook on Kentucky freezes for farmers
and industrialists in Kentucky. Application of the data for agricultural interests is
treated at some length in the text. However, owing to the variability of industrial .
applications, these are mentioned only briefly on the premise that each industrialist
will be able to make his own application after the data are made available.
Spring and fall freezes are discussed in the text without mathematical or
statistical references on the assumption that most users are not likely to be interest- "
ed in mathematical derivations. For those interested, a bibliography is presented,
which contains sufficient reference material for documentation of the mathematical
background and soundness of the data presented. _
2

 I CONTENTS
Page
Introduction ............................. 5
` Freezing Temperatures and Plant Injury .............. 5 l
Freeze Classification ........................ 6
_ Favorable Conditions for Late Spring and Early Fall
Freeze ............................ 6
Source of Freeze Data ....................... 7
Explanation of Charts and Graphs ................. 8
Proper Use of this Data ....................... 9
Literature Cited ........................... 31
I 3

 
 LATE SPRING AND EARLY FALL FREEZES IN KENTUCKY
A- By Doyle Cook*
During the course of producing and marketing a profitable crop, the Kentucky
farmer has many hazards to deal with. One of the more important of these is, in
_ many instances, the risk of a damaging late frost or freeze in the spring and the risk
of a similar occurrence early in the fall. The truck farmer and orchardist are possi-
bly the most concerned with late spring and early fall freezes, but those who raise
some of the other more basic field crops such as cotton, corn, and tobacco sometimes
experience losses from unusually late freezing temperatures in spring. Likewise,
some late-maturing crops are also subject to damage by early fall freezing.
Since it is not always economically feasible to attempt to prevent the occurrence
of freezing temperatures in Kentucky fields and orchards, the next best thing is for the
farmer to plan and carry out his operations in such a way that his risk of damage will
be minimized. To do this it is necessary that he have information regarding the chance
or risk of freezing temperatures occurring in his part of the state after certain dates
_ in the spring and before certain dates in the fall. In some instances the length of the
growing season (the number of days between spring and fall freezes) determines the
economic feasibility of growing certain crops. The purpose of this study is to provide
the needed information concerning growing season and spring and fall freezing temper-
atures.
While the effect of sub-freezing temperatures on agriculture is of greater im-
portance, other significant interests such as construction, sale of seasonal items,
maintenance, heating, etc. , are affected to varying degrees.
The data in this pamphlet will inform the farmer, .home gardener, or orchardist
of the risks involved in early and late planting dates. These data should also be of
considerable value to certain industries in planning their respective operations.
· FREE ZING TEMPERATURES AND PLANT INJURY
It is generally agreed that injury to plants by freezing usually occurs when ice
crystals form within the plant tissues as a result of the freezing of the liquid in these
tissues. The freezing point of the liquid within the plant tissues varies greatly from
one plant to another and at different stages of development in the same plant. For this
_ reason there is a wide range in the temperature necessary to cause major damage to
growing plants. A temperature of freezing or slightly below will damage and possibly
kill very tender young plants but may have little adverse effect on more hardy, older
plants. Peach blossoms usually sustain little or no damage from a temperature of 30OF,
but serious damage may be expected with temperatures of 280F or below.
The duration of sub-freezing conditions is also an important factor in deter-
mining the extent of damage. lf the temperature remains below freezing for only a
short time, the damage may be negligible, while the same temperature over a period
of several hours could cause major damage.
Other factors that seemingly have a definite bearing on the amount of damage are
the temperatures present prior to the sub-freezing conditions, the suddenness of the
*Agricultural Meteorologist, Weather Bureau Airport Station, Louisville, Ky.
5

 temperature drop, and the wind speed. Several days of unseasonably warm weather
in early spring will often cause perennials to come out of dormancy prematurely and
become especially vulnerable to later freezing temperatures. Some types of plants
seem to be more susceptible to sudden freezes than to slowly dropping temperatures,
even though the same temperature may be reached in both cases. Drying winds, that
often accompany unusually severe outbreaks of cold weather, can add considerably to -
the damage.
The appearance of frost has commonly been associated with plant injury. Frost
is defined as the deposit of atmospheric moisture in the form of feathery ice crystals
on the ground or other surfaces, the temperature of which has fallen to 320F or lower.
Frost and dew are actually the same phenomenon, the only difference being that, in
the case of frost, the atmospheric moisture (water vapor) is deposited as ice crystals
instead of water droplets when the temperature of the air at the surface is at the
freezing point (320) or below. Dew or frost can be expected to occur only on clear
still nights. Since sub-freezing temperatures can and often do occur without the
appearance of frost, plant injury may occur in the absence of frost. For this reason,
air temperature is usually considered a better criterion for measuring the extent of
plant injury than the occurrence or non—occurrence of frost.
Very little specific information is available regarding the critical temperatures
for truck-garden crops. However, these crops are usually classified as tender, semi-
hardy, or hardy according to their ability to withstand low temperatures. Tender plants
will be damaged or killed by any temperature of freezing or below. Included in this _
group are watermelons, tomatoes, beans, and peppers. Temperatures slightly below
freezing are necessary to cause injury to semi—hardy plants. This group includes I _
carrots, lettuce, and celery. Hardy plants, which include cabbage, turnips, and garden .
peas, can withstand a fairly hard freeze without being killed. ·
FREEZE CLASSIFICATION
Freezes have been classified (4)*as light, moderate, or severe, as follows:
Light Freeze — temperatures 290 through 320
Moderate Freeze - temperatures 250 through 280
Severe Freeze - temperatures 240 or lower
It may be generally assumed that a light freeze will kill only the tenderest plants;
a moderate freeze will damage most plants to some extent, with heavy damage to fruit
blossoms and tender and semi—hardy plants; and a severe freeze will cause heavy dam-
age to most plants.
The purpose of this study is to present the probability that a freeze of at least
light, moderate, or severe intensity will occur after any date in the spring or before
any date in the fall.
FAVORABLE CONDITIONS FOR LATE SPRING AND EARLY FALL FREEZE
In late spring and early fall, temperatures in Kentucky usually remain above
freezing during the daylight hours and drop below freezing only at night. The last spring
freeze and the first fall freeze are more likely to occur on clear nights. Fair skies
permit the invisible heat waves, which are constantly leaving the earth‘s surface, to ·
pass more readily through the atmosphere. Each day a large amount of heat is lost from
the soil and plant surfaces by these heat waves.
 
* Numbers in parentheses refer to "Literature Cited, " p. 31.
6

 Because of the loss of heat, the earth's surface and the air near the surface
_ gradually become cooler on clear nights. The cooling process continues throughout
the night, with the lowest temperature occurring near sunrise. If the air is sufficient-
ly cool to start with, below—freezing temperatures will result.
When skies are overcast, the heat waves leaving the earth are reflected back
to the earth in some degree by the cloud cover and not so much heat is lost from the
_ surface. By this means, cloud cover prevents the temperatures near the surface from
falling so much, and freezing temperatures are not so likely to occur.
Wind may prevent the occurrence of freezing temperatures. On clear nights,
the temperature near the surface is cooler than the temperature a few feet above the
surface. The wind mixes the cooler air near the surface with warmer air above it.
_ This prevents cool air from accumulating within the plant cover, and sub-freezing _
temperatures are not likely to be observed. One should remember that the speed of -
the wind is usually less at night than during the day. An afternoon breeze may die down
during the night, and the mechanism for mixing the air may be lost. Producers of
freeze—sensitive plants should observe the wind after sundown before concluding that
the wind will prevent the occurrence of sub-freezing temperatures.
L The first prerequisite for late spring and early fall freeze is the presence of
cool Canadian air over Kentucky. If this air is so cool that its temperature is below
freezing upon reaching Kentucky, freeze damage will result even if the skies are cloudy
and the wind is blowing.
Producers may be warned of such conditions by the Weather Bureau forecasts
which will, in such cases, warn of "much cooler temperatures," "cold wave, " "becoming
cooler tonight, " "frost or freeze tonight," etc.
Not all locations in an area have the same temperature. On clear, calm nights,
temperatures are cooler in valleys than on the adjacent hillsides. Cool air, being more
dense than warm air, moves down the slopes of hills, accumulating in the valleys. Ow-
ing to the cooler valley temperatures, orchard and garden sites should be located on
~ slopes. Because of less erosion and higher soil fertility, truck gardens are often
placed in bottom lands; but farmers should recognize that there is a greater risk of
late spring and early fall freeze damage in the valley locations.
SOURCE OF FREEZE DATA
Fifty—three U. S. Weather Bureau climatological stations in Kentucky and eight
» stations immediately adjacent to Kentucky in neighboring states are the source of the
temperature data presented in the accompanying charts.
National Weather Records Center Project No. 1013 provided summarized freeze
data for 38 of the stations used. Freeze data for 23 stations were summarized by the
author from data available in the Annual Summary of Climatological Data for Kentucky,
published by the U. S. Weather Bureau.
Statistical tools, such as mean freeze day numbers, variance of freeze dates,
and total years of record were provided in these summaries for the last spring occur-
rence of 320 or lower, 280 or lower, and 240 or lower, as well as similar data for
the first occurrence in the fall. With the devices thus provided, it was possible to use
the Normal Curve Area Tables for computation of the risk of freeze for any given date.
7

 H. C.S. Thom and R. H. Shaw (1) have shown that the statistical variance of freeze
data for individual stations are not significantly different within the area of one state.
This makes it possible to analyze the variances for a state as a whole and to derive
a single variance and standard deviation to use for all stations within a state. The
variances and standard deviations for station freeze date, as computed, for Kentucky .
are as follows:
Table 1 — Variances and Standard Deviations I A
 
Standard
Freeze Threshold Variances Deviations
Spring V
320 or lower 133. 06 11. 54
280 or lower 163. 93 12. 80 ‘
240 or lower 208. 56 14. 44
Fall I
320 or lower 134. 79 11. 61
280 or lower 123. 17 11. 10
240 or lower 161. 36 12. 70
These standard deviations may be used with any individual station mean to
obtain the freeze hazard distribution for that station.
The location of the stations from which data were used is shown in Fig. 10, 1
page 28. The thermometers were exposed in standard Weather Bureau instrument ·
shelters, usually at a height of 5 to 6 feet above the ground. Temperatures thus
obtained should be representative of the general area. However, on clear, still,
nights temperatures may vary greatly in short distances. Under such conditions, the
lowest values will usually be found in low places and near the ground. Also tempera-
tures in and near the larger cities will usually average somewhat higher than those in
surrounding areas.
EXPLANATION OF CHARTS AND GRAPHS
For the vast majority of operations there is a quick method for determining
the statistical chance that the last spring freeze will occur later than a certain date
for any given area of the state. While not strictly correct, mathematically speaking,
the following method is considered adequate for normal farm or industrial operations.
Figures 1, 2, and 3 show the average dates of the last occurrence in the spring
of 320 or lower, 280 or lower, and 240 or lower in Kentucky. Accompanying each fig-
ure, and indicated by a corresponding number, is a graph based on the data in Table 1.
Graph 1 is a curve to determine the date after which there is a given risk of 320 or
lower occurring; Graph 2 is the same sort of curve for 280 or lower; and Graph3 is
the curve for the last occurrence of 240 or lower.
In the past, considerable attention has been given to the average date of the last
killing freeze in the spring and the first killing freeze in the fall. This has been used
by many as an indication of the progress of the spring and fall seasons. From an '
operational point of view, little knowledge is gained by this figure, since it represents
only the halfway point. There remains a 50-50 chance of freeze occurring after the
8

 average date in the spring or before the average date in the fall. This certainly is a
greater risk than most farmers can afford to take.
After determining, from Fig. 1, the mean date of the last occurrence of 320
or lower in the spring, Graph 1 may be used to determine the statistical risk that
the last temperature of 320 or lower will occur on any date.
For example: If a farmer near Paducah (McCracken county) wishes to post-
pone planting a particular crop until the statistical chance that a temperature of 320
or lower will occur is less than 30 percent, he would first read from Fig. 1 that the
average date of the last occurrence of 320 or lower in Paducah is April 3. Using
Graph 1 at the 30 percent line and reading from the curve downward, he would read
+6. By adding 6 to April 3, we see that there is a 30 percent chance that a tempera- ~
ture of 320 or lower will occur in Paducah after April 9. After this date the farmer
should have 320 or lower only 3 years out of 10; in 7 out of 10 years, the last occur-
rence of 320 or lower will have taken place before this date. Conversely, if this
same farmer needed to know what the statistical risk of temperatures 320 or below
was after March 24, he need only enter the bottom of Graph 1 at the -10 line (March
24 comes 10 days before April 3, the mean freeze date . . . thus the -10), and pro-
ceed upward to the curve and read 80 percent. There is an 80 percent chance that
the last occurrence of 320 or lower will take place in Paducah after March 24. If,
owing to the contemplated crop, the farmer's interest was in the last occurrence of
280 or lower or 240 or lower, this data may be obtained from Fig. 2 or 3 and their
accompanying graphs.
Freezes in the fall damage late-maturing crops. At times farmers are forced,
by unfavorable weather in the spring, to plant crops later than normal. When this
occurs, they often desire some knowledge of the probability that a crop will mature
before a damaging freeze occurs. These data are obtained from Figs. 4, 5, and 6
and their accompanying graphs in the same manner used to obtain the data for the
last spring freeze.
The relative shortness of the growing season, "freeze-free days," influences
greatly the choice and use of many crops and varieties by the farmer.
Figure 7 shows the average number of days between the last occurrence of 320
or lower in the spring and the first occurrence of 320 or lower in the fall. Figures
8 and 9 show similar data for the 280 and 240 thresholds. The accompanying graphs 0
for each figure will enable farmers to obtain data concerning the probability that the
‘ growing season will be any given length. These graphs show the risk that the number
of "freeze-free days" will be less than any given amount.
For instance, at Henderson where the average number of days between the last
occurrence of 320 or lower in the spring and the first occurrence of 320 or lower in
the fall is 197, there is a 10 percent risk that the number will be less than 176 (the
average or mean minus 21 as read from Graph 7). In 9 out of 10 years, the number
of days between the last occurrence of 320 or lower in the spring and the first occur-
rence of 320 or lower in the fall should be greater than 176. In 1 year out of 10,
there should be a shorter growing season in Henderson.
PROPER USE OF THIS DATA
Data in this study should be used only to determine the probability that spring
and fall freezes will occur before or after certain dates. These data are obviously
not forecasts that freezes in any given year will or will not take place by a certain date.
9

 100  ,,.
._______Yl____  
70 innnnnimnnnn
65  ,
.______YI_____
S5 %______H_____
50  ‘
n-_--_. ---_-_
M5  ‘
 Wn-__Il______ 1
 .,-“--YI_-__-_
 .,__-_'l-______
n __________
Q 20 +15 +10 +5 Mean -5 -10 -15 -20 -25 -30
Date
Graph 1. —Curve t0 determine the date after which there is a given risk 0f 320
or lower in the spring.
10

  €£?’?? *  
¥€££§$%§ 
 "¥§%饕(  A
   @    
\` § '     ¤  
r    E E
 YE?}    
ۤ?     
Wg   
i`  QQ I   §  
¢ *1 1 ll  * Q
i is  
1 S6 ;  

 100 —-——
% -- - ---- ... 1
9% ---------  --
W --------"--
M-------- ]---  
Mun----.]-- 1
7% ““---“----
70% -:----m--
65% ------' I----
60% ------d-----
5% - --K-----
   -==-' —-- 
[+5% -I-dn--“
M Imi----
35%  H-----
ml   I A  
m—-“-;------
10% V »
 --4- ----- -
  ‘  
 ·' 0 +25 +20 +15 +10 +5 Mean -5 -10 -15 -20 -25 -30
Date
Graph 2. -Curve to determine the date after which there is a given risk 0f 280
0r 10wer in the spring.
12

 C" ; \-X ·°`\
  €@*%*\  2
       
-;’ ghgw IQ  ia  
    €&¢g§Q¤(      
     
   vc,  
¥    
   "$*:¢ ,  x.  
 §*=$’     
i   @1   Q
i   Q
I  
§

 100% r
-----------
9%  
90% *
85%------- fl--  
75%----:-- --
70% ---- --H -
------ H -
65% · _
 ----- A- - -
--- -- -
- - - -
50%
- --- -- -
/45% A
-----71-- --
M---- '
35%   --  
 ----H - - - r
25,; 
5 M---- -
2%  A----- --
 QZ--- - -
*1 + + • + J + • +  Mean —  - O — 5 -20 -25 — O
Date
Graph 3. —Curve t0 determine the date after which there is a given risk 0f 240
0r 10wer in the spring.
14

 ig :5/}\·-\__\\ 3% Ig
§ A U      G Eb
 §§?       
     
     `J A    
 !%;§&\§;‘  
 'l;
{ T =
I I   ~.
I · \
I I ~·· —.
I     I Q \·.
§ ;{'L,H       \·
5 2 ·   ·   L- °
ima \ :I /#   =~?§—~.   I
- V,-ac; 5-·"··*.\ 2 I ,· 2 s<   >\,>;VV\__ V
’I"’· ·~——r- RV is - Z   -   = ,<;’1— ’» \I
.AfQr‘;__7$         »-¤;\. ii) n.
kw/· gw, E @‘-s> ~#~i¢.·=,¤.#2,g;  
I) ·.<-wh`,. {Q E @   {JO`   51     [(*7
  4 @Jy¤V rz g,“·s.,,.r Vi;   WL; ¤.V   VV
JFS     aw YI B-, .»‘<   xg;/` 1 I =I
{SQ 2 FM ?* £#— F ii 2*: ¤=·# J I· `3 ‘-
V {,2,7 ·xé&9w:g\d J y ; A g l\_ .2; \
’~}f"" », C ‘·~·I - O Q`~». ,\.` V A I Z
gif: — gn; pggp {ag Le ,>l;,#·._ *·?I”“~.A~ ·E»./I
Jjci \.:¢ A {‘ F ·‘ g /’; >/ ~ G] Z VJI >   .
¤ —F,,. K‘   WZ ‘ /·”I"`:I 2 ·v;%q S I +@ —< . / \ ~E
‘»“·—— V g>‘¥’.·;~.;·;·“¤ \   Msg 2 éjm   .4   J acm K \ ~ Vi}
  I     :2/  %-Q-V\_h—»Vx¤q G { ;$
; 4 ~ —4 T z *7 _ ‘.,»1—4mV;’_· L ·· 2 ..
T   U   ir. 2         7Y-777*-»(7—§_'4;;}':—·—l`;’ , .¤~_-i"
      I      "’?·j».»? 3"?‘}#iL··—;iiQ§C g   Q A
I/'.~\`* ,.»;~ Air"); EE Iz »»» r· V :i*;v—-ff ’·* .
I' {W · VJ; f ·~ - ; I ·\.~\L \ D 6 I .¤ O
\~<*` ·"‘f-)"Y`/*‘.¤\Z. = ‘E’ (K. `¤     O Q)
·<»“""°&  &_/-{»;\?§°Zejx 55/ 2 22 I C: `p~Z·~.E   2 gf
{E  QU! £`<.\ = I : :A`>~\; IN “_,_VI;¤ CD CL
`“*—¢-J ¤> ¤
/I /; { ·-X ;_/___ _·< Jv____ L _•_;
Vm/C   - X ·‘ -6.1 )._I=· sr _ I VV »¤
M Iyé   `{ ; O T/i;==x¤ (E., M L , " ·§ m
Gy 92% at   1;,.; ·- a¤
·;; i  LN   \ N, ¤ I3 V
    { ¤‘ " `é=`® ·— B Q
I>*   9 Ur J   °  I`:`*I ¤>
  V____*__r` V I ({- —\ ;{T_\_§//I   g YQ
H Ibn 5-·;~·_ >¤ ex D; Y z\.;¤_ S-I <1>
I {JT? {YE €Li•`¥““·{§“ /·/I ·· 4   E §
U   =   K iA=I,V»·i—#°¤( I Ze; I ¤ m `
l.l II {—`TV¥> ` ‘ 7:" _~rI,__§ O
D   I VIE; /V?¤‘%   \‘*·—-—ZZg —{~_ ¤¤>‘ E?
M I  /·    II E I `= EI, -¤
Fl O "II I lj T   \` · LIJ—·?   Lis MFLI O _ (D E
Q1     V:._       '   Ii;;_\·`i I U `
7 $0II IQVIIIQEI   *\\.   { I   it I O I é
I ——~*VV_ * JI I IQ   I ?* \ ‘E   Q I   I s 2
` `* I1] I 1»Vx;i’\§ \? ,:\VI,.f;_ *
 — {V V IgT;(n_f g \_ V_;§.y—J° j 5 X `·—I ·
V · . r/wt T       Q Z  
»-4 I V EI‘:~$g”"' »     ~   ¤ [J • Lu
\;`¤   _·7r\,,_u'_ T} \. _,-/( W
I     ,VA»%_jiV.`:LII/_ v"` E
I I `I»¤·’"_VV_I~·~V>?’     2 I
I     ‘
·——{`_ I   · é: V
 »-~ I   W *: I
VQ I I ' 7   _ /';·/;;;;°Ii"T.‘T~f.__ :
I  
    VI
  _ fj ·; · V
28

 STATIONS FROM WHICH TEMPERATURE RECORDS WERE USED
1. Hickman 25. Dale Hollow, Tenn. 49. Heidelberg _
2. Cairo, Ill. 26. Summer Shade 50. Mt. Sterling
3. Paducah 27. Wolf Creek 51. Farmers
4. Mayfield 28. Greensburg 52. Maysville
5. Murray 29. Bardstown 53. Ashland
6. Dover, Tenn. 30. Clermont 54. Grayson U
7. Princeton 31. Louisville 55. West Liberty
8. Fords Ferry 32. Anchorage 56. Inez
9. Mt. Vernon, Ind. 33. Shelbyville 57. Dewey Dam
10. Henderson 34. Madison, Ind. 58. Pikeville
11. Madisonville 35. Markland, Ind. 59. Hindman
12. Greenville 36. Covington 60. Benham
13. Hopkinsville 37. Williamstown 61. Baxter
14. Clarksville, Tenn. 38. Cynthiana
15 . Russellville 3 9. Frankfort
16. Beaver Dam 40. Lexington
17. Owensboro 41. Dix Dam
18. Tell City, Ind. 42. Richmond
I 19 . Irvington 43. Berea
20. St. John 44. Somerset
21 . Leitchfield 45. London
22. Mammoth Cave 46. Williamsburg
23. Bowling Green 47. Middlesboro
24. Scottsville 48. Manchester
29

 Daily forecasts are issued by the U.S. Weather Bureau for all of Kentucky.
These forecasts contain detailed temperature forecasts and should be used as day-
to—day guides in planning for protection against late spring or early fall freezes.
For longer-range planning purposes, five—day forecasts are issued three times each
week on Monday, Wednesday, and Friday. These provide, in more general terms,
temperature and precipitation forecasts for the up-coming five-day period and may
be used to plan farming operations for that time. e
Twice each month, on the first and fifteenth, 30-day forecasts are issued
which give an estimate of the average rainfall and temperature for the next 30 days.
It should be kept in mind that nearly always there are changes in the weather
every few days, sometimes of a sudden and severe nature. Notice of these changes
is widely disseminated in the daily forecasts and in occasional warnings. _
This study of Kentucky freeze data should prevent farmers from taking undue
risk with regard to freeze—susceptible crops. However, if a farmer, owing to his _
own economic situation, is willing to take a chance on an early planting or late har-
vest date, this study will at least answer his question, "what are my chances of
success ?"
30

 LITERATURE CITED
(1) Thom, H. C. S. and Shaw, R. H. "Climatological Analysis of .
Freeze Data for Iowa." U. S. Dept. Comm. , Monthly Weather
Review, 86 (7), July 1958, p. 251.
(2) Carter, Horace S. , "Late Spring and Early Fall Freezes in
Georgia." Ga. Agr. Exp. Sta. Bul. n.s. 41. 1957.
(3) Frederick, R. H. , Johnson, E. C. , and MacDonald, H. A. ,
"Spring and Fall Freezing Temperatures in New York State."
N.Y. State Col. of Agr., Cornell, Misc. Bul. 33, 1959. ~
(4) Decker, Wayne L. , "Late Spring and Early Fall Killing Freezes
in Missouri." Climatic Atlas of Missouri No. 2, Mo. Agr. Exp.
Sta. Bul. 649, 1955.
(5) Anon. U.S. Dept. Comm. Weather Bur. Project 1013, "Freeze
Data," National Weather Records Center, Asheville, N. C.
(6) Whitney, D. Ransom. Elements of Mathematical Statistics. The
Dryden Press, New York, 1959.
(7) Cook, Doyle. Unpublished freeze data. U.S. Dept. Comm.
Weather Bureau, Louisville, Ky.
l 31

 4M—8—64