f^ltlr:

m

flfS

J^>^w

:^ J*^ >•■ ii M )«

%

-y**»^

* .**A/vi

\A^^j%^.nKmKmhhi

;u - ".-,'; i";.r

-Oc ^^•^«'

T^.^n

.'i

J'iltnii-,,

'Ski^^^nr^'"'

V n l^f ^ -'. . (-iiv\< .

A PHYSIOLO^^IC iL STUDY OF THE CLII.TA-^IO CONDITIONS OF MARYLAND AS I.fEASUHSD BY PLAN'^ ^^OW^H.

( A second contribution from data obtai nad under the auspices of the Maryland State 'leather Service, in 1914.)

DISSERTA-^ION

Submitted to the Board of University Studies of the Johns Hopkins University in conformity with the requirements for the degree of Doctor of Philosophy.

By F. Merrill Hildebrandt

M

Bartimore , June , 1917*

A PHYSIOLC^I^AL STUDY 0^ ^H'^ OLIMA-^IC CONDITIONS OF ilARYLMD AS I.^EASURED BY PLAN'^ OHO'^TH.

( A second contribution from data obtained

under the auspices of the Maryland State

Weather Service, in 1914.)

F. Merrill Hildebrandt.

OUTLINE.

Pages.

I. Introduction 1 to 32

A. Plan of exoeriment, location of

stations etc. 1 to 4

B. The use of standard plants 4 to 6

C. Experimental methods. 6 to 9

1. The plant used.

2. The soil ani its preparation,

3. The watering device.

D. The climatic measurements. 9 to 26

1. Introductory measurements used etc.

2. Treatment of climatic meas- urements.

a. Temperature - a general dis- cussion of the methods of handling temperature and a description of the method used.

b. Light - description of method used in giving an index to light.

(!. Evaporation - '^he evaporation index,

E. The plant measurements, 27

F, '^he method of handling the data used

in this study. 28 to 32

1, Reduction to daily averages per plant.

2. Reduction of daily averages to r relative values,

II Presentation and interpretation of data, 32 to 97

A. Introductory, including an expla- nation of the graphs and tal)les. 52 to 37

B. 'Vhe two-week climatic data, 37 to 48

1. Introductory

2. General and detailed considera- tion of the temperature, light, and evaporation graphs.

3. The relation between light and evaporation and the relative variability of the three cli- matic conditions.

C. The two-week plant data 49 to 66

1. Introductory

2. Outline of plan of di scussion

3. The relations "between the plant measurements

4. The seasonal marches of the plant measurements at the various stations oomoared.

5. Correlation of the plant and cli- matio data. The general plan of cor- relation and its application to the plant and climatic values for the individual sta1;ions.

D. The four-week climatic data. 67 to 69

E. The four-week plant data. 69 to 89

1. Introductory.

2. Outline of plan of discussion.

3. The relations between the plant measurements.

4. C!orrelation of plant and cli- matic data.

P. The covered stations 69 to 96

1. Description of covered sta- tions.

5. The behavior of the plants of the covered stations.

3. The possible explanation of this behavior and its bearing on interpretation of the fa-^ts of the experiment. G. The forest station. 96 to 97

1. Description of the forest sta- tion.

2. The behavior of the plants of the forest statioi'..

3. The possible explanation of this

and )tS 'h&c^r\'r^c'| behavior^,on the interpretation

of the experiment results.

Ill '''he soy-bean as a standard plant. 97 to 104

A. Introd-uctory. 97- to 98

B. The plant measurements con- sidered as reaii.ngs of the

standard olant. 98 to 99

0. The plant producing power of

the climatic complexes of the

various stations. 99 to 100

D. A comparison of the seasonal

plant prodmding power of the stations. 100 to 1^1

H. The internal conditions of

the standard plant. 101 to 104

IV. Conclusions. 105 to 107

-T^ables f Plates I to VIII) 108 to 115

Txraphs ( Plates IX to XIII) 116 to 149

AC KNO'-XED ai.TEN T 3 .

The study dealt with in this paper was sug- gested by Dr. B. S. Livingston and carried out un- der his direction. The writer 7;ishes to express his indebtedness to Dr. Livingston for great assis- tance rendered in c^^rrying cut the study and for helpful criticisrn of the manuscript. The writer also wishes to thank Tr. H. E. Pulling for valuable suggestions made during the course of the study.

iinitoDuc^^iOK.

During the summer of 191'^ an elaborate invest J gition was rmder taken by the Maryland State Weather Service in ooop- eration with the T^ahoritory of Plant T>':---;j olo^-y of the ITohns Hopkinr University, v/ith the object of ascertaining some of the relations between olimatio conditions and the growth of certain -'-.lants at different st^.-^ions i.n Maryland. T)etailed information as to the growth of the plants used is of coxirse necessary, as is 'also corresponding knowledge of those envi- ronmental -'onrl i ti ons that are considered as cli>a'*"5o. The plant records were secured in this case by gi'owing cultur;^s of certain plants in the environmental conditions to be studied an-i. noting the growth made during definite i^ei^iods of time. In order that a corresponding series of .".easure-. Eents of some of the environnental conditions might be avail- able ■''or comcarison with therae growth measuremf^rtp +-'-•- n';:i+nres were located 8,t certain of the regular U. S. 'Veather Bureau obRervaticn stations at various places in the sti+e. '^he general plan of the stuly and a detailed consideration c-"' the methods used has already been printed by McLean^, -A-ho

X' McLean, ?. T., A preliminary stiidy of clima'^ic conditions ^n Maryland, as related to plant growth. :^hysiol. Res. 2: 129-808.1917.

did all of the field work personally. "-.e original data dealt with in the present naper are taken from the records obtajne-"! by McLean ann it will be necessary to give here only

SO much description cf the v/ays i" whi^'n t^e^e '-neasur ■^r = -ts were obtained a? is needed to render tlsm intelligble. This description is mainly taken from McLean's paper, which deals with the growth .1' soy-bean -lants, but for onl^- -^^-'O of the stations, Easton and Oakland. The present paper gives the m.ain results for soy-bean slants, for all of the stations, together with some attempts at interpretation. This stij-^^- has been carried out partly through financial aid furnished by the Maryland State V/eather Service.

'^he stations employed were Oakldnd, Ohewsville, Monrovia, College Park, Baltimore, Darlington, Coleman, Saston, and Princess Anne. One station, Oakland, is in the Allegheny plateau. Four stations are 'n the piedmont plateau, one fChev/sville) in the Hagers-^own valley, two CDarlington and Monrovia) in the hilly country north and west of Baltimore, and one (Baltimore) at the Itywer edge of the plateau near Chesapeake bay. Pour stations. College Park, Coleman, 3aston, and Princess Anne, are in the coastal plain. Coleman, 3aston and Princess Anne are eas . . Chesapeake bay, -'-^t^ ^nii c~o Park is west of it and much farther inland, near tiie line of demarcation between th" coastal plain and the piedmont plateau. All ' C-. *-.-•-; o>"_s except Oaklnn'' ere ^t comparatively" l'"^" elevatioiis.-less than 310 meters (1000 feet) above sea-level. Oakland has an elevation of 775 meters (2500 feet), '^he geographical distribution (rj^-~ ^ig, 1) c:'' +-hese stations is such that considerable differences in climatic conditions exist5between them.

3

At each of the nine places employed in this investiga- tion, a series of cultures was grown in the onen with no cov- ering other than a screen of large-meshed wire ne-^ting to nre- vent injury to the plants, "^hese have been termed the ex- posed stations. In addition, at Oakland, Baltimore, and Easton, a series of cultures was grown under glazed cold frame sash supoortel horizontally 1 meter (3.3 feet) above the surface of the soil. These have been termed the covered stations. They were placel within several meters (6.5 feet) of the enclosures containing the plants of the exposed stations and were subjected to the same climatic conditions as the ex- posed plants except in so far as these conditions were raodi- fied by the glazed cold frame sash. At Baltimore a series of cultures was grown in the woods near the Laboratory of Plant Physiology of the Johns Hopkins University. This has been term- ed the Baltimore Forest Station. These plants like those of the exposed stations had no covering other than a protective wire screen. Ov;ing to their locatioii, they were, of course, subjected to a set of climatic conditions quite different from those acting on the exposed and covered plants at Baltimore. The I^'orest Station at Baltimore wa^ distant about 150 meters (490 feet) from the exposed and covered stations. There are thus plant data available from 13 series of cultures in all, each series having been exposed to a different set of envi- ronmental conditions throughout the season.

7/30

tTW

JgOO

TffOO^

?ig. 1. lipp or Maryland, showing locations o" stations employed for soy-bern cultures fnc climatic observe t ions . (i->.fter iv.cLeenj

Ihe environr.ental conditions to which the cultures of this experiment were exposed were so controlled that the plants might be regsrued as standard piants for the measurement of climatic conditions in accordance with e suggestion made by Livingston and :..cLean^''' . Since the problem of expressing

V Livingston, B. li., and i.xLean, F. ^- A living climatolo- gical instrument. Science, n, s. Ac: c62-c6'c. 1916.

plant growtri in terms of climatic conditions that control it

5

is rencierea exceedingly complex ty the nurrber cf these conditjons sna their continue! verif-tion, fs well ss by the ch^ngin/x inter- n&l conditions of the plant itself, e detailed analysis of the control of plant growth in terrr.s of effective climatic conditions is very oifficult, but, as Livingston and iicLean suggest, the rate of growth *of any plant is itself an expression of the sum total ail the effects of the external conditions acting during the growtii period, so that a standard ^olant rright be eir.ployed as &n eutorsa tically weighting, integrating, and recording in- str'acent foj' the cor.parative neasurenent of growth conditions as these act on plants. I'.ius tc;veral environments i^.sy be measured ana compareu in ternns of their several capacities for producing grovith in the standard plant. This method of rr.easur in*"' environ- ment in terr.s of plant growth can be appliec. only when it rcay be assumeu that all the standard plants are alike at the begin- nings of the several periods of exposure. In the present study the requirement just stated was fulfillea by employing the seed as the starting point for the plants of the various cultures. It was apparent that if the cultures were always started from the seed the plants migh" be considered as more nerrly alike at the beginning of the several culture periods than woula have been the case if an attempt haa been made to obtain. like plants in any other phase of their uevelcpment. The internE 1 conditions of tiie plants change continually, however, during growth, and no two of the cultures were the same at the end of the culture periods. Tiiis phase of th« problem will receive attention later.

As in other problems in which a num.ber of" conditions enter into the control of a process, the' relat ions between

6

conditions and process rate are more easily detected tfee smaller is the nuinber of conditions involved, and conditions may be left out of consideration if they are the same in sev- eral experiments. Just as the internal conditjor.s of the standard plant are left out of the arguniemt by the simple device of having them all alike at the beginning of the ex- ■posure period f the instrument being set at zero of its scale, in the v/ords of livings-^on and Tv'cLean) , so selected ones of the surroundings may be left out of consideration by having them alike throughout all of the periods. According to this -nrin- ciple all of tlie environmental conditions that acted on the plants belovf the soil r^urface were kept practically constant at all times and at all stations. Assuming that the artificial control of the subterranean environmental co^.ditions T/as thus practically constant, -^iie differences observed in the grov;th of the standard plants vv'ere taken to be related almost entire- ly to the aerial conditions of the surroundings. These are the ones referred to by McLean afc climatic, and this terir; v/ill be used '>vith the same meaning in the present paper. To a.ccom- plish this control of the subterranean conditions, the soil was always the same at the beginning of all cultures and its moisture content was generally kept approximately the same throughout all culture periods, by means of the Livingston auto- irrigator. "^he arrangement and its operation have been de- Gcribed by I'^o'Lea.n. and will receive some attention below.

The grov/th rates of the plants were meastirXsd and com- pared in terir.- "'' their size ani weight. Zr?.cih cultue consist- ed of six plants grown for a leriod o:'' four weeks from the se=d. Cultures were started approximately every two weeks

7

during the growing season, at eaohi of the stati ons emplopied, and growth measurements were made after about two vreeks and again after about a month. The plants were hari'ested at the end of the longer period.

While several different plant species ?,-ere employed throughout the experimental work, the present paper deals only with the data obtained from soy-bean. A variety of this olant called "Peking", v/as used. The seed was oT' pure str-.in obtained from the 1913 crop of the Maryland Agricultural Exoerimert Station. All the seeds were first treated with carbon bisulphide vapor for one week, to destrogr insects, after which they were placed in paraffined paper cylinders with tight- fitting covers and stored until ready for use.

The same kind of soil was used in all of the plant cul- tures, at al 1 stations. It w'lS a rather light soil obtained from an untilled field near College Pari?, Maryland, and was of the type classified as ITorfolk Sand by Bonsteel. ""he top-

Bonsteel, J. A., "^he sojls o.^' "^rince C^eorfrels '^oimty. Pub. Maryland ^eo?.ogioal Survey. Baltimore, 1911.

soil was removed from a small area of the field to a depth of 15 cm. , and thoroughly mixed and sifted. It was then -nlaced in cloth sacks and shipped to the various stations where it was stored in air-dry condition. in covered, water-tight, gal- vanized iron cylinders, until needed for use in the cultures. The soil cnntainers for the cultures were ordinary "6 inch", porous clay flower-pots , in form like the frusticxiin of a cone, being smaller at the bottmm, and of a cubiri capacity of approx- imately 1980 cc.

8

In order to secure uniform soil conditions in ^'m various cultures, it was necessary not only that the soil should be of the same character in all of them but also that it should be brought into the same physical condition for the beginning of all cultures, furthermore, it was desirable that this physical condition be such that it would be retained during the growth periods of the plants. To put the soil into a state of aggregation to be least altered by varying wee,ther conditions (especially heavy rains which "oack the vsoil more or less) it was saturated with water immediately after being put into the pots. This was accomplished by plunging the filled pots into a bucket of water and allowing them to remain submerged until air bubbles ceased to rise. The pots were then allowed to drain.

The soil moisture .in the cultures was maintained always above a certain minimum by means of auto-irri gators. This

^Livingston, B. 3., A n.ethod for controlling plant mois- ture. Plant V/orld. 11: 39-40.1908.

device, as here used, consisted of two cylindrical porous clay Gu-os (o'' the regular form supplied by the Plant Y/orld) connected with each other and with a water reservcir by glass tubes in the -'''-> r-r^ of an inverted J. The cups were placed ver- tically in th oot, their rubber- stoppered tops level with the soil surface, and Tvere so arranged as to supply w,- ter to the soil against a pressure of 35 cm., or somewhat raor-- , of a water column. The moistxire content of the soil was thus maintained so that it was never less than about 10 to 13

9

per cent, on the "basis of dry weijfeht. This particular soil with this water content was rather too wet than too dyy for the best growth of the plants here studied.

After preparing the v>ots and arranging the watering de- vices the pots were then allowed to remain fallow for about two weeks before planting. Thus the soil v/as fully drained after the preliminary saturation £.nd had settled into a con- dition somewhat approaching that of structure equilibrium before the seeds wore planted. The seeds were planted 2.5 cm. deep, six seeds in each pot. Oare was taken to space them un- iformily and to place them about equally distant from the auto- irrigator cups and from the sides of the pots so that all should have, as nearly as possible under the conditions of the exper- iment, the same soil moisture conditions, "^hen the plants v;ere removed from a pot fabout six weeks after that pot had been filled) the soil was discarded and fresh soil from the stored supply Y/as used in refilling for the next folloT-rin^ culture .

The weather observations taken by the cooperative ob- servers at each of the sta'^ion here employed consisted of dailv readings of maximum and minimum thermometers, daily ocular observations of cloudiness, daily measurements of rainfall and general notes as to storm.s, winds, etc. In addition to these records of the weather observers evaporation was ;neasured by means of Livingston standardized cylindrical oorous cuns with non-rain absorbing mountings. Of the fiv^ sets of climatic

J^

Livingston, B. S. , A rotating rabl§ for standardizing

porous cup atraometers. Plant V,'orld l.>. i:)' -162.1912. Idem,

10

Atmoraetry and the oorous cup atmonieter . I b j d . 18:21-30, 51-74, 9.5--111, 1^..3-149.lyl.)

■-"h'^cr''' -ti ons nentionefl aToove, only t'lree rfi 11 "be conslfiered here, narael:^ those of temperature, light, and evaporation. As was pointed out by Mclean, rainfall f^howed little or no relation to the growth of these -slants, since the soil mois- ture of the cultures was always kept sufficiently high' ("by the auto-irrigators) for the needs of the plants. The main influence usually thought of as exerted by rain uoon slants is of course an indirect one; the rain does not affect the

plants but it alters the soil moisture condition and *-he al- teration thus brought about influences the vSn.pply of v;ater avail- able to thfe plant roots. :ainfall data will therefore not be dealt with in thife paper. Also, the miscellaneous cliraatol- ogical observations reported bv -^he weather observers "'ill not be considered since none of tiiem have been found to bear any relation to the growth of these plants.

The mass of climatic data, instruinantal In the case of evaporation and temperature, and observational in the case of light, can obviously not be compared with plant growth until it is simplified in -"ome way. '"he process of p i-riol i ''in^ition here adapted involves two main steps. The first of these con- sists in bringing together the daily observations into t^-o- week and four-we'-^k -roups corresoonding to the two anfi fox<r- Yieek growing periods, i :' way in which to do this, -'^i' instance, is to average the daily readings over the two-week or the four- v/eek period, ■^hus securing an avera^^'e ."iaily value for the

11

climatic rneas-are.iients in question. The second step consists in weighting avarage values in such a way as to furnish a second series of values that express the climatic conditions as they affect the lolants. It is clear, for example, that the readings of a therrcoxeter do not express the effectiveness of tenipersture to accelerate or retard ;f)lant growth. It there- fore becoraes desirable to re-clace the -actual thermometer read- ings Tjy a series of weighted values, more or less d,i, re-^.tly pro- portional to the temperature effect upon the growth of the ■plants. O-^ing to lack of information of a quantitative nature as to the relation het'.veen olant growth and environmental tem- perature, this can be accomplished only in a tentative and approximate v:aj at the present time, llo sttempts have yet been made to derive such weighted values to represent the effectiveness, for lolant growth, of any other climatic condi- tinn.

The consideration of temperature , light, and evapora- tion now to be g:iven will show how the original climatological data have been grouped for comparison with the plant growth measurements and haw the average values have been handle 1 in ""he present study.

'tempera •ture .

The temperature data used in this study were all ob- tained from maximum and minimum thermometers read daily at sunset, "^he daily mean temperature was determined by averag- ing- each days maximum and minimum readings. Uhe maximum and minimum temperatures from which the daily means were de- termined were secured from the ^^ublished monthly re-oorts of the U. S. 'Veathei- Bureau.

vT'assig,,0. L., Climatological Data, Maryland and Delaware Section, numbers from i:ay to November inc. 1914. U. 3. Y^eather Bureau.

It'cI.ean ha,? discussel some of the possible ways in w'njr>h daily maximum and minimum temperature data may be trea"^ed in order to obtain weighted values that may represent -tempera- ture effect upon plant growth rs.tes. He points out tha" tem- perature values as shown by a thermometer do not -show a linear proportionality to plant growth. If thermometer readings ex- pressed, even in an approximate way, the effect of tempe- rature on growth, s-.ich a rela-''ion could nnly be true up to the optimum temperature, since beyond this point increased terapera*:ure results in decreased growth. It would, therefore, be desirable to replace each thermometer reading by an index representing the efect of that particular temperature on plant growth. Three ways of -lolng -^his, all of which have been con- sidered by McLean may receive brief mention here.

One way of expressing temperature, which has been \ised in ecological studies, maybe called the remainder sumjTiaticn

u

method. Tiiis Is based on the supposition that the growth activities of niany or nioit plants stop when the temperature falls to about 40" Fahrenheit. Above this temperature, growth increases with increased temperature, to an optimum. For convenience, the growth ra^e for 40) *> ?. may be consid- ered as unit:/; them it should be 2 for 41'', 5 for 44^*, 20 for oQ**, etc. If we subtract 39 degrees from anj'' given tempe- rature, then, the remainder will represent, according to this method, the efficiency of the temperature in question for producing growth. A total efficiency value for any period of time, such as the four-week growth periods of the cultures of the J nvestigati on here considered, might be obtained by sub- tracting 39 from each daily mean temperature and summing the remainders for the period. This raefchod has frequently been used in ecological studies 7/here it was desired to obtain approximate expressions of temperature values in terms of their efficiency to produce plant growth.

Another method of weighting temperature values for the purpose before us, and one that has an apparently more ration- al basis, was suggested by Livingston and Livingston.'^ They

V Livin?:ston, B. B. and Livingston, 9, J., '^emperatur" coefficients in plant geography and climatology. Bot. Taz. 55:349-375.1913.

proposed a series of temperature efficiency indices based

on the Van't Hof f-Arrhenius law, whcih states that the velocity

of many chemical reactions approxim.ately doubles with a rise

in the temperature of ten degrees C!entigrade (16° F. ) . These authors assurae the growth rate to be unity for a temperature of A"^" ?. amd derive a series of values representing tempera- ture efficiencies fir hi^-her temperature. In using these. ex- ponential indices the assumption is made, as the authors have pointei out, tha-^ the plant processes whose rela'^ion to tem- "oerature is under investigation follow the chemical principle upon v.'hich the indices are Lased. When this scheme is use:l, the efficiency value for any temperature is represented by

Tj (\r>

the value of the exponential index that corresponds to^^ tem- perature value itself. Assumi ng t"hie growth ra-^e to be unity for a temperature fo 40'* ?. , it should be 1.21 for a terar)e- rature of 45°, 2.0 for 58°, etc.

Since most of the tempera txires with which we ^-ave to deal are belov/ the optimum for plant growth, since temperature and the growth rate appear to be related in an approximately linear manner between 40° ana the optimum, f about 32" C.) and since both the exponental and remainder series of index values increase in a practically linear way throughout this range, both of the methods Just considered give temperature effi- ciency numbers that appear to be approximately proportional to plant growth as it is influenced by t^mperatxire . It is of co-rse obvious that neither of these methods can properly express efficiencies for temperatures above the optimum since the.7 give numbers which continue to increase with increasing temperature 7:hile growth increases with increasing tempera- ture up to the optimum and then decreases .with higher tempe- rature. Also, both these methods appear generally to give

15

approximately proportional results for ordinary growing tem- peratures. This fact has been noted by Livingston and Livingston and again by Stevens and it Is also obvious from

f

Stevens, Ileil 3., Influence of t emperature on the growth

of Endothia Parasitica. Am. Jour. Bot. 1: 2, 112-118.1917 Idem, Influence of certain climatic faotO'S on the develoo- ment of Endothia parasitica. Ibid. 4:1, 1-33.1917.

the climatic data given by ITcLean.

A third method of expressing temr)erature ae ' -^ effects plant growth has been more recently suggested by Livingston . From the results of Lehenbauer's experiments on maize seedlings.

^

V Livingston, B. E. , Physiological temperature Indices for the study of pXant growth in relation to climatic conditions. Physiol. Res. 1^: 8, 399-420.1915.

V Lehenbauer, P. A., Growth of maize seedlings in relation to temperature. Phyfeiol. Res. I ;247-288.1914.

Livingston derived a series of coefficients giving the effi- ciencies of various temperatures in terms of the growth of this plant.. He has called these "physiological temperature indices". The growth rates upon which the index values are based are those shown by the seedlings when exposed for 12 hours to a maintained temperature, the other conditions of the ex-^eriment being approximately the same for all experi- ments. It is suggested that t e coefficients thus derived

16

from the growth of maize under controlled conditions, with different maintained temperatures may possibly express a general relation between plant growth and .temperature and may thus be applicable to plants growing under other conditions. The graph of these physiological indices, as rela'^ed to tem- perature is of course the graph of mr.ize seedling growth as so related, and it exhibits the same direction of slope between low temperature and the optimum graphs of temperature efficien- cies as iorived by the other t7/o methods, but for this portion of the temperature range the slope of thp graph of physiologi- cal indices is generally .:;teeper than that of the graph of remainder indices, the latter graph having a much steeper slope than that of the exponential indices. This is shown by Livingston and also by Stevens, in the papers cited above. Since they are derived from the actual growth rates :;f a plant, the physiological temperature indices appear to have a more rational basis trian either the remainder or the exponential indices. For this reason, and for others that will appear below, the physiological indices are used in this study for expressing the tempera■^nre as it aff'^cts the growth of the plants. Two other series of temperature values sire presented in the talles of this paper but neither has been found to be as satisfactory for exioressir.g this climatic condition as a:'e . the physiological indices. ":hese two other temperature values are fl) the average daily mean temperature for each period, in decrees Fahrenheit and (2) the remainder summation index for eahh period. In the case of temperature, as in the case of light and evaporation, the value given for each period

represents the averajre daily value. All of the data hnre treated, both plant and climatic, have been reduced to daily rates, for reasons which will be given below.

McLean has pointed out three ways in which we may use the daily raaximtun and minimum temperature record and a tempe- rature coefficient, such as the Livingston Physiological index, to get average daily temperature ef ilciencies for growth per- iods. (1) We may add "^he maximum imd minimum for each da"^, divide by two to get the mean temperature for the day, and average the daily means thus obtained to ^et an average daily mean for the period in question. (This gives the aeries of numbers shown in the tables of climatic data, line 5.) The physiological index corresponding to the average daily mean for the period may then be taken as the temperatxire efficiency for the period., (Z) We may sum the physiological indices Corresponding to each of the daily means, divide this sum by the number of days in the period and thus get an average daily index to represent the temperature efficiency for the period. (3) Lastly, we may average the indices corresponding to the maximum and minimum for each day, thus obtaining an avera^-e daily index, add these average daily indices, and divide the sum by the number of daj^'S in the neriod as was done in the proceeding case to get an average daily index, '"he first method takes account only of the variations between periods, the second involves the differences between periods and the interdiurnal variations, while the third takes account of both differences between periods and the interdiurnal varia- tions and also invor'''Q5s the daily range of temperature. Only

18

the second of these threee methods has been employefc 3.n the present paper.

To bhow the reason for using the physiological tempe- rature sumnaticn index in thip study, rather than the suiri- mation of the remainder or exponatial indices, it will he necessary to anticipate somev7hat the discussion tha"^ is to follow, The three climatic conditions ( temperature, evapo- ration and light) shov/ a definite seasonal mar?h for each of the stations employed in the investigation. The temperature r.i ses from low values in thp Rin-ing to a midsummer maximum which is follov/ed hy a subsequent fall to low autumnal values. Cn the other hand, the values representing both light and evap- oration decrease, in general, throughout the season. Tf, now, a generalized curve of plant growth be drawn, plotted against the time of year, and employing average values to represent all the stations together, sxich a curve follows the seasonal march of the tempera +"ure and shows only secondary variations due to the effect of the other climatic conditions. The growth of the plants is thus determined mainly by temperature. Obviously, also, the seasonal march of the temperature values must show the same general form of curve no matter wha"^ scheme is used in expressing teraperf-ture efficiency. In view of these facts, and in considera-^ion of the general comparative purpose of the present study a method should be used for ex- pressing temperature efficiency, that gives a seasonal march of the efficiency values in accord with the corresponding march of plant growth. Of the three methods mentioned, the physiol- ogical efficiency index fulfill'^ -^his requirement best, and this has accordingly been selected for use throughout the entire

study. An examination of the plant and. olimatic graphs fto "be considered later) ^;ill f=hOT)7 that the plant values for most of the stations rise above the temperature efficiency values in the middle of the season, and fall below theqi at its end. This is probably due in part to the effect of ligh^ and evaporation as will be brought out below, in ^.^-' "i i scus- sion of the plant data, but it may also be related to an in- adequacy of the temperature efficiency values to represent the actual effectiveness of temperature in growth control. It appears to be at least suggested that the actual tempe- rature efficiency "alues for these soy-bean ilants increase m.ore rapidly with increase in the temperatue itself, ior the range here encoiintered (between 40*^ add So""?. ) , than lo the physiological index values derived from Lehenbauer's study of mSulize seedlings. This whole question deserves much more epxf^^rimental study. It is a surprising fact that we have available only a single thoroughgoing investigation of the relati on of temperature to the growth of higher plants, in spite of the fact t'nat the primary importance of the tempe- rature control of growth is obvious to every observer and has long been qualitatively appreciate!. A com.parison, for any of the stations employed, of the range of growth values for the plants with the remainder sura^^atlon values for temperature (which ar- practically equivalent to the exponential summa- tion values in this study) and with the physiological summa- tion indices will furnish evidence for the verification of these sta*:ements. '^he graphs 6'f the physiological sumniati on indices of temperature efficiency shofi much steeper slopes than do the corresponding graphs derived from the other two

kinds of temperature indices mentioned above, however, so that the physiological indices are evidently more suitable to represent temperature efficiencies than are either of the other kinds.

21

Light. The only records of light conditions tha^ were availahle

for all of the stations of this study 'jrere the daily ocular

estimates of cloudiness furnished by the cocnera*:ive weather

observers. To rakke use of these estimates it was, of course,

first necessary to bring these daily percentages of clear sky

together for each culture period, so as to derive for each

period a single value that might be taken to represent the

Intensity of the light condition. The method employed to ac-

complish this is p esented in the next following paragraphs.

^ The presentation of this method is here ^practically the same as that previously published. See; Hildebrandt, ?.F. , A methcd for approximating sunshine intensity from ocular ob- servations of cloudiness. Johns Hopkins Un'v. Circ. March, 1917.

The total heat equivalent cf the actual sunshine for any given period at a given st ."^ion is primarily a function of three terras: (1) the maximum possible number of hours of sunshine (determined by latitude and season); (?.) the me-m intensity of full sunshine for the period and station, which may be expressed In terms of heat received per unit of a horizontal surface; (3) the condition of the sky, whether overcast, partly overcast or clear, '"he daily values for the first two of these temms vary In a regular manner throughout the year at an:/ given place, and the ones for the thJi rd term are roughly stated in the observer's records, as Just mention- ed.

■^he first two terms are combine.i ' -•^. ^he or-linates of •'■he

22

graph given ty Kimballv farr the maximum possible total ra- diation received per day at Moimt Weather , Virginia. Since this station is at about the sp.me latitude as the stations

Kimball, Herbert H. , ^he total raiia-ion received on a horizontal surf'ace from the sun and sky at Mount Weather, Monthly Weather Rev. 42; ^74-487. 1914. (See especially fig. 8, p. 484).

here dealt with, the ordinate from Kimball's graph may be taken as approximate measures of the total maximum possible light intensities for the corresponding dates for all of the Maryland stations, '"hese values represent the total amount of heat received from the sun and sky on niear days at Mount Weather, in gram-calories per square centimeter of a horir.ontally exposed surface. The method of using this graph and the weather observer's reports, for estimating sunshime intensity for any station and period, will be best shown by an example. Suppose it is desired to estimate the average daily sunshine intensity for some sta'"ion in the general region of Mount Weather, for the first v;eek of August, '^he average ordinate value for this week is first obtained from Kimball's graph. For leriods as short as a week or two this may be done by averaging the values for the first and last days of the period, since the curve may be taken as a straight line for such short intervals, "^rom the report of the weather observer at the ;flace in question, the number of

2'3

clear, partly cloudy, and cloudy days is next determined for

the days August 1 to August 7, inclusive, and some arbitrary

v;eighting is given to eahh kind of day. This was done in

the oresent instance by recarding days reported "clear" as i;7hole days

of sunshine, those reported "partly cloudy" as half days of sunshine, and those reported "cloudy" as without any sunshine. The same scheme of weighting must of course be adhered to in all the estimates used for comparative purposes in any dis- cussion. By summing these weighted daily values a number is obtained that represents the equivalent number of clear days for the period considered. Suppose, in the example selected, that this equivalent number of clear days 's 3.5 v;hich is 0.5 of the total number of days in the week period. The latter value Tiay be termed "the coefficient of clear weather". By multiplying the average daily intensity value for clear days, as already obtained, by this^oeff icient of clear weather a value is secured tha. t may be taken as a rough approximation of the average daily sunshine index for the v/eek.

While it is certain that solar radiation affects plants in other ways than through its heating effect, it is no less certain that by far the greater part of the energy of susishine absorbed by plants is converted into heat (largely as latent heat of vaporization), and it seems probable that the other effects produced upon the plant may be more or less proportion- al to the total energy equivalent of su shine. This method of deriving sunshine indices, although it is to be taken as only a rough approximation, has been shown, as a matter of -^'act, to give quantities rather definitely correlated with the plant groivth values in this study. It has been found, for instance, that the amount of dry substance produced

24

per unit of le a.f area in young soj-tej^n plants decreases from the beginning to thn end of the growing season, in a manner that generally parallels a corresponding fall in the light intensity values as determined in the manner dascribed above.

Bvap oration. Evaporation was measured in thf^s^ sti;dies by means of cylindrioal porous cup atmo/aeters located so as to have the same local exposure as the plant cultures. The atmometer mountings were provided with -mercury valves so arranged as to prevent the entrance oi' rain. They were read at intervals ifif' about two weeks, the da"*"es of reading being the same as those on -''h'oh observations were made on the plants. After every reading each atmometer cup was re- moved and. replaced by another that had Just been standard- ized. The use'i cup was subsequently restandardized so as to detect any change in the coefficient of the cup conse- quent upon its exposure. 'j7h.en the re standardization ed a change in the coefficient, the m.ean of the original co- efficient and the coefficient found upon restandardization was employed to reduce the reading to the Livingston cylin- drical standard. The evaoorati on readings should therefore be directly comparable to other mei^surements related to the same standard.

As has been pointed out by Livingstonj the porous cup atmometer is somewhat similar to plant foliage ih the way in which its evaporati^ng surface is exposed to the sur- roundings. It may therefore be supposed tha" the transpi- ration from the plants for any period should be approximately proportional to the evaporation from the atmometer, except in so far as the transpiration rates may be influenced by conditions within the plant. The work of Briggs and Shantz indicates that evaporation from small open oans or porous cups is inflnenced by the same external conditions, and in

v'3.

^ 3ee their paper, cited on p. -15

26

about the c -.s is plant tra.iio_;_^J.Xtt!.J v; li, J j' the compari- son is made for periods of a day or more. Of coturse the two rates do not vary proportionally vrithin the day period, since the internal conditions of the plant exhibit a neculiar daily march, but m th such details this study does not need

^ r-^ ......

Ti;,n i-o"^ ^ -^ i" -^ - '- -P =1

o r' ^ -v» ^- -V-

soil moisture and to evaporation. Carnegie Inst, Wash, Pub, 50. 1906.

to deal. It has been supposed therefore, tha-^ tiae effective- ness of the external conditions to influence the transpira- tion rates from the plants of this stiidy was approximately measured by the corresponding evaporation rates from the ^tmometer. The atraometer re dings have been reduced. In every case, to mean daily rates for the E-reek and 4-week periods taken as indices of the evaporating power of the air as it affected the transpiration rates of the plants.

27

Plant IJeasurements.

T^he first plant measurements v/ere taken after anoroxl- raately tv/o weelcs of growth fro;n the seed. At tiiat time the length of eaoh leaflet from tip to junction of blade and netiole T;as determined, as v.as also the greates'' vridth of each leaflet, measured at right angles to the long axis. The height of each plant was also measured, from the soil surface to the base of the terminal bud. At the end of approxi- mately four weeks of growth, the height measurement was re- peated, after which the plants were cut off at the soil sur- face, and the dry weight of tops were subsequently determined. Before drying, photographic prints were prepared of the fresh leaves. Bj means of these leaf-prints the leaf area (one side) was afterwards determined planimeterically. All linear measurements were made to the mearest millimeter, ^eal measure- ments to the nearest 0.1 sq. cm., and weight measurements to the nea.rest 0.01 gram.

2H

The reduotion ant and climatic measurements to

average 3 for the period and to relative values.

As has been sta-^ed previously, each culture originally comprising six plants was observed after about two -weeks of grov;th from the seed and again after about four weeks from the seed, and the coBsecutive cultures 7/ere started at inter- vals of about 2 weeks. In many cases, however, the nximber of plants from which records were actually taken was less than six fit ^as never less t'-.an 3 and W:.s usually 4 or 5 in such cases) , on account of observab"'-e injury due to other conditions than the ones here studied, such as insect attack, etc. All plant data are^theref ore , stated as averages per plant. Also, in many cases, the length of the preiod varies slightly from 14 days for the two-week rjeriods and from S8 days for the four-week periods, and the averages per plant have consequently been expressed as mean daily values for the res- pective period. This method renders the plant measurements for the different periods more strictly comparable. It should be noted, however, that tfee growing periods "-ere 14 and .26 days long in the malority of cases, and that variations in the X length of the culture periods were slight. Considering the 2-week and 4-week plant values as measures of the results

of plant processes acting through the periods, the mean daily values represent mean daily increments or process rates, and they will be termed "daily incremants", for their res-nective periods, in the discussion that follov/s. Thus, for a. l,.j.i.u

Iw

10 cm. high at the end of a 13 day period, j^ cm. is re- garded as the mean daily increment of increase in height, etc..

2\i

for that period. Letting the v?ord grov|rth represent the ^aV"

t'f^'ul'":/' ;J33 to -^ich the given Irirxd of measurerrent re-

fers fas increase in height, increase in leaf area, etc.), these may he spoken as growth increments,

a,

The mean dily growth jr.crements and t'le mean dail'^ climatic values for the respective periods have been express- ed in terms of the corresponding average of all the periods considered , for all exposed stations. This prccedure rentiers all the values directly comparaoie . To obtain this unit for any kind of value, all of the corresponding values fas all 2-week daily mean increments in height, for example, for all exposed stations) 7-'ere sumnied, and the sum was divided by the number of values summed. Then each individual value was di- vided by the unit thus obtained. The data are expressed as these ratio values, which will be termed relative values in the following discussion. The absolute magnitude of the tmtt thus used for exi^ressiri^ ea?H Virri r>f value is«of course. not important; it is essential only that all comparative values be expressed in terms of the same xmit. The unit here employ- ed represents in every case simply the average of all similar quantities that are used In the present study. If another station had been employed, or if the season had been longer or shorter at any station, the values of these comparative units would have been different. The magnitudes of these units thus depend to some extent upon the climatic condi- tions encountered at the various s ;.c. x^ns in the simmer of 1914, to some extent upon the number and location of the stations, to sc-ae extent upon the nature of the soil used in this investigation, and to some extent upon the physiological

30

nature of the soy-bean plant. The values of these various ■units are all given in table I. To avoid ; " ' ~ -^'-■^e rela- tive values secured as noted above, have all been multiplied by 100, and are thus ^iven in the tables fplates I-VITI).

Table !_ for all exposed stations.

cli::a7ig

Avej'age daily physiological temperature index = 55.39

Average da'ly evaporation index = 16.2 cc.

Average daily sunshine intensity = 442 calories per sq.cm.

PLADI'^: 2-week periods.

Average daily increment in stem height per plant =3.56 ran. Average daily increment in leaf-pro'-"'"'"-^ ^er plant =112 sq.iri'-:

PLANT; 4 -week periods.

Average daily increment in stem height per plant = 3.20 mm.

Average ddily increment in leaf area per plant = 122 sq, mm.

Average daily increment in dry v^eight per plant — 6.29 mg.

""he use of these relative "-.l-Les simplifies the plotting of the graphs upon which the interpretations of such a study as this so largely depend. It also renders possible a direct comparison between the values for an>/ t o cultures irrespective of their date or sta-'"ion. furthermore, it is possible to tell from, the magnitude of the rela'^ive value for any culturs- the extent to rhich the plant or climatic measurement under consideration der>arts from the mean of that measurement for

n

all the cultures of the study.

To obtain the original or absolute plant or climatic value from the relative value as given in the tables of this paper, it is necessary only to reverse the arithmetical procedure by which the relative value was derived, j'or ex- ample, suppose it was desired to get thejactual mean daily rate of increase in leaf-area ver plant for the four-week period beginning Aug. 5, and for the station at Ooleman. The relative value given in the table is I'^B. The first operation is to divide by 100, which gives 1.08 as the true relative value; The average daily in'^reT»ent in leaf arv^a per plant, for the period and stati '-n in question, was therefore i.06 times the value of the common unit employed for this process of increase in leaf area. IJulti plying thus unit value Q22 sq. mm.) as given in talble I by 1.08 gives 132. 0 sq. mm. as the average daily increment required. To obtain the average total leaf area per plant at tlie end of the period in ques- tion, we multiply 132,0 sq. mm. by the number of days in the period f 28 in this case) and get 3698 sq. mm. Since there were 5 plants ;neasured in this culture, the total leaf area for the entire culture at the end of the period is obtained by multiplying 3698 sq. ram. by 5, which gives 18490 sq,mm. or 184.9 sq. cm., which is the actual areal value detexmined from the prints of these leaves. All of the original ab- solute values may be obtained from the relative ^nes in a similar manner. It is of coitrse evident from the above de- scription of the m^anner in whj ch the relative values have been derived that they are proportional to the corresponding

absolute valuss. In all subsequent discussion, when plant

32

audi cliEnatie valuf ^ referreil tc , it mil be iinderstcod

that these are the relative values rather than the absolute ones.

Presentation and physiological interpretation of lata.

Introductory. The plant and clijnatic data \vill no\'j "be presented and described. The discussion will be devoted in part to descriptions of the growth changes observed in the plants at the various stations, and for the various periods at each station, in part to corresponding descriptions of the climatic values, and in part to some attempts to correlate these two sets of data. Owing to the coraplezity of the prob- lem and to the number and variety of the data to be dealt with, it has been found necessary to depart frequently from a general lorical order and to treat matters that are of second- ary importance at greater length tlran may appear necessary from a more restricted rjoint df view. Such interpretatlonsas are here attempted are of interest partly for their own sake, but rnainly because of the bearing they may have on the gen- eral problem of the use of standard plants for the comparative integration of effective climatic complexes. It must be re- membered that the general project the results of which are flere dealt with was planne'i primarily with a view of making a first trial in the use o-^ standard olants in this way, and that such correlations between plant growth and the conditions of the surroundings here rendered apparent are to be consid- ered as 0-^' secondary importance to the main nurpose. "?hese discussions \'d. 11 be presented more in the form, of a running narrative, with digressions at many points, than ia ideally

3;i

aesirr-ble, out th«i nevvness of this kind of study rnd the f?ct thpt th" fundementsl principles anci even the terns to be used heve net yet been aeveloped meke jsnything aoproachinf^ p true lorrical sequence quite impossible.

The various kinas of dsta to be considered will be brought forward in groups corresponding to their sources. The two-week plant data ana the two-week climatic data will first be presented followeu by some ettem.pts to correlate the two groups from the view-point of plant physiology. Then the four-week plant and climptic dpta and their physiological correlation will be pre- sented. These topics will be followed by a special discussion of the aata for the covereu stations end a similar treatm.ent of the oatf for the forest strtion at Baltimore.

u

presentation of the Data.

Two aietiiods of presenting the data of the experiment have been emplo7/ed in this paper. The relative ntun'bers, derived as n6ted above, have been p\'rr--n in tables, together "^i ff-' ^'- <= ^^^es of the first and last days of each culture period and other in- formation including the length of the neriod, the number of plants in the culture, etc. Also, a set of graphs is -oresented ^'nrmr- ing grpahically certain parts of the information given in the tables.

The t- bles of plant and climatic data for the various stations employed are shov/n in plates I-VIII and will be de- scribed below. "Inhere are, in addition, several tables given in Jtihe text ^vhich will be described elp-^whp.re. Plates T-*^ITt contain 26 tables in all. Nine of these, shown in plates I-ITI inclusive, contain the plant and climatic data for the two« week culture periods for the exposed stations; ^''-r-^ ^ shown in plates IV-VI inclusive, give the data for the &ur-week cul- ture periods for the es^posed st-.tions; four, shoLvn in plate VII, give the data "for th® two-and -FciT''---ve-> .-.-■!+::-■;•.. periods for the covered stations at Oakla d and Baltiniore; and four, shown in plate VITI give the da' a for the tv/O-and four-week culture periods for +^'-e covered station at ■Sastoii ^i-^-'f -rnv fi-e Baltimore Forest Station.

Tn all of the tables, '-' "irst line giv=;s the name of the pla-ce <?. *• ^'•'"'jch the .. '' ■,■*■'-'; e lengt

ture periods (whether two-or four-week) , and the character of the exposure of the plants, that is, whether the sta'-ion 's ex-

6 'J gives also the dates of the beginning and end of e^ ilture

period/ at the head of the column in which the data of that particular culture period are ^^^2*4 ily.ced. The -econd line of each table gives the culture numbers. "^hese numbers being given the various cultures for convenl«=nce in reference, When several kinds of stations occur at vu'z place, cultures of the same niuaber cover approxiroatel^r the same time period. For in- stance, at Baltimore, there is an exposed station, a covered station, and a forest station. The two-week culttireSnumbered 6, for the exposed, coverecl , "and forest stations, at 'Baltimore each grew from August 20 to September 3. In some cases, cul- tures limiM*^ ' the same number for the exposed and c'-"'--"' stations show a difference of a day in the lengths of their respecti^'^e loeriods owing to the fact that •■ t was impossible to take L.f;c.t^arer.ient s on both the exposed ajia oov-si ^ 1 ^iltrnt^j on the same day. The' third line of all the tables gives the length of each culture period in days, this number being ob- tained for cii'xj ueriod by siibtracting the dates given at the head of the column containing the data for that period. The fourth line Jn all the ta'les gives the number of plants used in obtaining the plant aeasurements . In all the tables, a dash appearing in place of a rela ive value indicates that the data necessary for calculating this relative value is lacking. An asterisk Placed opposite a climatic Ixne^. shows tha"^ *"' " '"-^ex was not plotted in the grpahs fto be described latere . This is done in the case of most cultures v/here no plant data are available for comparison with t'-r climatic values. The last coluLin of each of the tables gives averages, for the station, of the plant and climatic values presented.

36

The tvvo v.eek tables fcr "-^"^ exposed stai;ions show In line 5 the remainder summation index for the respective cul- ture periods. As has "been noted, this is obtained by subtra^'-ting 39° from each d.cdlj mean i>i:.t summing the remainders for the period. Line 6 gives the average lanl^' relative physiological index determined in ths manner previously described. Line 7 gives the average daily mean temperatures. Line S t'^.e average daily rela-^ive evaporation index and line 9, the average daily relative sunshine in'ensity. Line 10 shows the values cf the average daily relative increment in stem height and line 11 the values of the average daily relative increment in leaf-product. The two-we<^k tables for the covered stations correspond to the two-v7eek tables for the exposed stations except that no tempe- rature cr sunshine data is available, and the tables thus con- tain only the relative evaporation indices and the two nlant measuren^ents. This is also true of the Baltimore Forest Station. The four-week tables correspond line for line irith the two- week ones except tiiat the four-week tabl.es show the aVerage daily relative increment in leaf area insteac^ "•'*' '^' ^' average daily relative incremant in leaf-product, ani a line is added to the four-week tables giving the average daily relative 5n- crem.ent in dry weight. 3ach four-v^eek value cf relative daily physiological temperature index, relative daily evapora- tion index, and relative daily sunshine intensity was obtained

by averaging the relative values of these climat"'-^ -"-^/^tors for

constituting the four-week period the two-week periccb^in questi.:. . "^he four-week peii od value

of the remainder summation index for any period was obtained by

adding the values of th' s '■'^.lex for the tv/o - two-^""^"- •-•oricds

making up the four-v/eek peri od uncler consideration. The average

3'/

daily mean tei^x^t; mature for the longer i^ericdvS was obtained by- taking the mean of the two average daily means for the two-v/eek period:-.

""'-- graphs, shown in plates IX to XIII, present graphically certain of the data given in the tables. In all of the graphs, the ordinates represent magnitudes of the plant and dljmatic ralative values and tii& abscissasrepresent. the t i..ne ■' -"' ':;t year. The ordinate scale is given at the left of each, set of graphs for convenience in reference, and the dat-^s of the be- ginningsof successive culture periods are shown on the base line. For the first two-week period for Oakland, thus, the ordinates shav the average daily relative values of the plant and cli- matic measurements for ^'— two-week period, beg, "a*^ P'^. '^'-^e 100 line of the ordinate- scale is the value, as was -oreviou-sly noted, of the seaso-al average for the state of each of the plant and climatic measurements taken. An explanat j on of the method of representing each of the measurements shown in the graphs is given in th-^ legend ' ^.late XIII. The t v^ o -week c 1 i ma 1 1 c_ d a t a . 3 ntrgguctory . The two-Vveek climatic aste consists of the rela- tive ueily averages cT the temperRture index, avrporstion index, anc. sunshine intensity 2<jv b series of consecutive periods exten- ding over the entire season, each period being about 14 deys long. These values thus furnish a continucua record of the ser.scn at each Etrrtio.'i. The four-v.eek periods, ho\iever, overlsp, each one in- ciuaing the last tv.o we-^s of the preceeding ana the first tv.G weeks of the following period. V.hile the climatic averages based on t: four-v.e«k apt? form f> smoother curve then uo the twc-weck values, small variations in the conditions arc to s great extent obscured by averaging the overlapping periods. The series of tvio-

6 a

week' values exhibit the march of the cljmatie conditions at each of the various stations in somewhat greater detail than do "'"'1" 'iiorresponding '^'=^rj'='s of four-v;eek values, sine t lie latter represent overlapping periods, and the former will therefore be made the basis for a sonaewhat detailed and comparative discussiO' of ^h^ climatic conditions for the various stations. Temperature will receive attention first and evapora,tion and light will afterv/ards be considered to- gether. In each case, ■^'le general characteristics i^conmon to nst or all of the stations) of the seasonal march of the condition considered will be brought out, after which attention v/ill be given to peculiarities of the v:.lues for individual stations.

Temperature conditions

The graphs represen'i'ing temperature conditions present the seasonal march, at each of the various stations, of the average daily relative physiological index. '^he most obvious general characteristic of this index value is tliat it is high in summer and low near the beginning and end of the season, for all stations. Graphs of similar form are obtained when daily means and remainder summations are correspondingly plotted but the discrepancy between the midsummer values and those for the beginning and end of the season is much raorejpronounded jn the graph of physiological index values There employed) than in either of the others. The second general characteristic of all the graphs of the plysiological index of temperature is that they possess two maxima, both of 7/hich have about the same magnitude.. The first occurs in the las'^ two weeks of

81^

July and the second in the last two weeks cf August, this sta'^ement being true for all the stations considered except- ing Oakland, fcr which station they both occur relatively early in the season, in the last two^aeks in Jiine and July respectively. A tiiird feature which is common to mo stf though not all) of these graphs lies in the fact that the up^irard slope is more gradual (before the occurrence of the high midsummer maxima) than is the downward slope fafter the occur- rence of the maxiiTH.) A generalized tem-oerature efficiency graph representing averages of the corredponding values for all of the sta'''io'.s is n;t symmetrical about the ordinate for its highest midsummer value; it slopes uow^-rd less rapidly than down7/ard. A foutth general characteristic of these graphs lies in the fact that the final low index values of the frost- less season are not ver«) different for the various sta'^ions. The following considerr.tion of the graphs for some of the in- dividual stations vlll serve to bring out the points mentioned above and will give opportunity to note exceptions to the generalized statements just made.

In regard to the forms and other characteristics of the temperature efficiency graphs, the nine stations studied may be grouped into five classes: fl) Ohewsville and Monrovia, fEi Baltimore, Darlington and Coleman, (3) Easton and Princess Anne, f4) College, and (5) Oakland. The last two stations i'o not appear to fit jnto any of the first three classes and they ar: not aliky au ':':i'^^ ma-jt aot be regarded as representing separa/'e classes. These five groups are dis- cussed in order below. It will be noted that the stations of groups 1, 2, and 3 are located near each other, and this

in

probally accounts for the fact that they show sj.mjlar graphs of the tempera tiire values.

Ghewsvllle .and Llonrovia. The graph of physiological tem- perature indices for Ghewsville sho\'7s all the characteristics mentioned as general throughout the series cf staions. It rises gradually during the first three periods, f period "be- ginning May 19 to oeriod heginning June 16) , '^hen drops slightly during the fourth period flieginning June 30) alter which it rises for the period beginning July 14 to a maximum of 149. The value for th'- 6th peric .ginning July 28) is relatively low fll2) , after which the maximum (145) occurs for the period beginning August 11. The index value in question then decreases rapidly during the next two -oeriods attaining a magnitude : f 46 for t'le 9th period fbeginning Sept. 6) and remaining low until the end of the frostless season. Monrovia has the same -ort of graph as Ghewsville, the maxima coning in the periods beginning ^uly 13 and August 10. The minimum relative value of the temperature index is 53 for the period beginning Sept. 21.

Baltimore, Darlington, and Coleman. At Baltimore, the physiological temperature values increase gradually to a maxi- mum of 162 for the period beginning July 9. The seconf maxi- mum comes in the preiod beginning August 6 after which there is a rela-^ively rapid decline of the index values to 62 for the period beginning September 3. The Da-'lington graph has its first maximxun in the first two weeks f July and its second in the two week period beginning August 7, and then falls off rapidly to a minimum cf 46 for t' « -^'vs*- '^eriod '-^i September.

41

The graph for Coleman shows a graiual rise, t^-'o maxlrna for the periods beginning July B and August 5 and a ray^i "i "^,11. The temperature record is imcomplete at this station and the lovi- values for the end of the season are therefore not avail- able .

Saston and Princess Anne. At Saston there is a graiUial rise to a maximum of 154 for the first period in July, f e second maximuj:! coiping in the period beginning August 17. The curve them falls to a minim\im of '.3 in t'-:^; i qpt no rid n-f the season. The Princess Anne curve sho?rs the two typical maxima in the periods beginning July 9 and Au^st 18 7/ith a minimum of 4.'3 for the last period of the growing season.

Oollege . The Qollege graph of physiological ""'^''-•es '9 unusual in showing a marked rise for the period beginning June 19, thus giving the graph three maxima (139, 148, and 143) for the periods beginni -g June 19, July 17, and August 14 respectively. The graph drops rapidly to a vlue of 50 for the period beginning September 25.

Oakland. The temperature index values for Oakland are all relatively low, being consj derably less than '^'•".e season- al average employed as unity. This graph shov/s two maxima, one for the last half of June and thr othnr for the last half of July. Each of these n-^ixima will be seen to occur about a month earlier than the corresponding ones for the other stations here studied. The Oakland graph is also unlike those for the other stations in that '^■'- downward slo-'^e i- ore gradual than in the oth~r cases.. Its final relative value is

43 for the 2-":eek period 'beginning Sept. IE which was the last -full period for this station, "before the occurrence of a killing frost. The p: •-:t outstanding characteristics of the Oakland season in respect to this temperature efficiency graph, as compared with the seasons at the other stations, are fl) general lo-- values of thfe ■•-■h-' biological temperature index, (2) short duration, owing to the occurrence of late spring and early fall frosts, and (3) the early occurrence of the maxini-;- . '^'■"•>- markeci l i -^ferences "between the Oakland graph and those for the other stations here dealt v;i th are no doubt largely due to the relatively high altitude of this station as compared v,'ith the altitudes of the others, bs has been mentioned by McLean in his comparative study of the Easton and Oakland seasons based on these same data,

Lenving the one Oakland cu"^ 'f accouiit, the other eight temperature efficiency graphs may be described as a single generalized graph, in the following general terms. Seginning ^ra. th Relative index value of aboi;*^ pn (for the first -lart of May) the graph rises to a maximum (about 150) fcr the first part of July, falls slightly and rises again to a second maximum of about the same value as the first^ for -'■^-^ first part of August, and .finally falls to a minimum value of about 50 for the last period of the frostless season. That the initial values are not ir.-nv • c. no doubt '■^■'•^ t'~ ■^'•-e fact that the various series of cultures were not started until some™ v/hat later than the beginning of the ^sasasBESEt^Sil:^ frostless season. foee McLean's paper cited above ^

i4 Evaporating power of the air and sunshine .

The two-week graphs of the values for atmospherj, c evaporating, ptjwer and sunshine will be treated together since '•"'^^ seasonal ""i-ar-'hes nf these two clinatio ccnfl i *:" npp. exhibit the same general characteristics. Three points .nay be noted in regard to the seasonal marches of these two climatic indices. (1) Both grapHq have, in -Rneral, a dovm- v/ard slope from the beginning to the end oi the season. (Z) In the majority of cases they agree in direction of slope, from v-pr^od to pp'r'od, throughout the sef^so-;^:. f5) The;' agree in having a primary maximum, with a very high value, fo? the early periods of the season and one or more secondary maxima, with lov/er values, f c - i^eriods ^ha'^ occur later. The second- ary maxima of the graphs for light and evaporation sometimes (but not always) coincide, as to time of occurrence, with a corresponding ms^i'nnTi or" the graph for temperatore efficiency. The following consideration of the individual station graphs for these two conditions may serve to bring out these points.

■Por Oakland the primary .-'H-v-im-am 'n th^ frraph of atmos- pheric evaporating power fl53) occurs for the first period (beginning May 27)). The value of the evaporation index then decreases steadily to a relative magni t-!i,i<^ nf 7Q fnr the +"irst two weeks in July, after which it increases to a maximum (104) which corresponds in time of occurrence (period beginning July 6) -^(^ ^he second maxim^im of "^hp graph 6f temperature efficiency. After passing this maximiim th6 graph descends again, to the lo- values 57 and 69 for the last two netiods (beginning August £7 and Sept. IE). The sunshine intensity

44

index varies from an initial value of 122 to a final value of 81, with maxima for the peri ods /beginning July 16 and August 14/. Inspection of these two graphs for OaKland shov/s that the direction of slope is the same from period to period, for the greater part of the season. ?or Ohewsville the two graphs agree in direction of slope throughout the entire season except betv/een the periods beginning August 25 and Sept. 8. Both curve graphs are approximately parallel to the graph of temperature efficiency for this station from the period beginning July 14 to the period beginning August 25 and both have a dovmward slope, in general, from fche beginning to the end of the season, and they agree in direction of slope from the period beginning June 15 to that beginning October 8. The maximum o-f the evap- oration graph for the period beginning July 27 corresponds to a secondary minimium in the gra'oh of temperature efficiencgr for College. The graph of the evaporating power of the air has a primary maximum for the second period fbeginning May 28) and a well-iparked secondary maximum for the period beginning July 22. IIo sunshine -^ t ^ >-■. -e available for this station. p-^r Baltimore the tv/o graphs in question agree in direction of slope up to the period beginning July 9 after which the evap- oration graph -ascends to a secondary maximum -n'} .-^h corresponds, in a very rough way, to the doiible maxiyima of the temperature efficiency graph. For Darlington the general statements made at the beginning of •<-'•■' s ■''iscussion hold t.hroughout the gre^+p-'- part of the season. The atraometric values for this station are relatively very low; all but tv.'o of them ar^: lower than the seasonal average employed as unity ' - '■'-■ --'^sent study.

45

Por Coleman the sunshjr.e record is inooniplete , but the two graphs generally agree in aireotion of slope so far as comparison is possible, excepting between the periods begin- ning July 17 and July 51, ?or Easton Bnd Princess Anne, the curves are typical. Svaporatiori data are lacking for the periods beginning .June 8 and Jiir.e ?3 at th"' latter station.

The conparatively close agreement, in their main characteristics, between the corresponding .graphs for sunshine and evaporation, for all the sta-^ions employed in this study, together with the fact that the latter graph exhibits no irell- defined relation to the grap'". of temperature efficiency appears to indicate that the rate at which water evaporaed from the

A

white cylindrical cups employed as atmometers in this inves- tigation was determined to a considerable extent by the amount of radiant energy absorbed by the cups and that the air tem- perature played a secondary rjart in the determination lT tills rate. The fact that the physiological temperature coeffi- cient is used for expressing temperature does not operate against this conclusion since, as has been -oreviously st.3ted, other methods of expressing temperature give curves which slope for the most part, in the s me direction as the curve of phjrsiologj cal temperr.tare indices. A large effect oi sunshine on evaporation, the sunshine being measured by a black bulb sunshine recorder, has been found by Briggs and 3hantz These authors v;ere able to calculate approximately the amount

V Griggs, li. J., and Shantz, H. L. , Hourly transpiration rate on clear daj^'s as determined by cyclic environmental factors. Jour. Agrlc. Res. 5:583-650.1916.

of evaporation froi:; a shallow 'blackeneri tank usj'^'- « formula which involved sunshjne intensity ani tir^e saturation deficit of thedtr, and in vrhich sunshine has a preponderating in- fluence. They also stnte that an approximate proportionalj ty exists "betv/een the loss from the tank and the loss from Livingston '^orous cup atmometers. While the cups and the tank respond in -■! * f f ersnt wajrs to the daily cycle of changes in the evaporating po-,';er of the air, a certain average ratio exists between the evaporation from the tank and the evap- oration from the cups. It 'vould therefore "be exoected from their work that e aporation from Livingston porous cups would he largely infl-;enGed by sunshine intensity, and that tem- perature would '^how a secondary influence on evaport'o-n as measured by these instruments. It must be remenbered, also, that the evaporation meaaurements of .this experiment were made in the plant enclosures, v.hile temperature was measured by thermometers locate 1 in a shelter abotit 1 l,/2 meters (5 feet) above the ground and often 4 or 5 meters fl5 feet) from the plant enclosures. This ma37 account in some measure for the apparent absence of any marked effect of temperature on the evaporating power of the atr as measured bv porous cup at- m.ometers.

It m^ be noted here that the temperature efficiency values for tiie various -tat ions here considered, exclusive of Oakland are much more nearly alike for any gi''*^^ two-WT?ek period +-han are the sunshime and evaporation values. The values of these three climatic indices ior the first two weeks of -June and for the first two v/eeks of August, for these

47

eight stations are given in table 2. Since the dates of ob- servation -jrercnot the same for all stations, these values have been approximated from the graphs, but they may be consid- ered as sufficiently accurate to illustrate the manner in which the data at hand supoort the conclusion Just stated.

Table 2.

Values o-'' the three climatic indices for the first two weeks in June and the first two we^lrs in August with ratios of highest to lowest values for each index.

Evaporation

Sunshine Temperatur'^ efficiency Cphysiological index)

1	-St 27;ks<	1st 2wks.	1st 2wks	1st 2v;]rSd	1st 2w>s	1st 2wks.
station, of June	of Aug.	0 f Ju :i e	of Aug.	of June	of Aug.
_ ;— j Ghewsvill 3	115	90	150	95	■ 105	1 ■ ' —•^^ 120
Monrovia	145	115	122	103	110	12 5
Gollege	156	147	-	-	103	133
Baltimore	109	110	92	77	102	152
Darlington	115	78	115	112	96	125
Coleman	^ .' '"	135	145	120	115	152
Saston	J_ ^;^	135	1G5	115	112	140
Princess	125	95	110	75	102	135
Anne
Ratio of
highest to
lowest	1.5	1.9	1.8	1.6	1.2	1.3
valTie in
a\70ve ser-
ies.
Average
for the
2 periods.	1.	70	1	.70	1	. 25

If the highest value given for ea'-.h of the three indices and for each of the two periods be divided by its lovrest value th

' 48

ratios oresentel in the rest to the last line of the talole are obtained. iSach ratio represents the iTiagnitude of th-: range of variationy of the olinatic index that it represents for the eight stations in question. The average ratio value for these periods is given in the last li.ne . It thus appears that the geographical range of variation in the temperature efficiency index is marl-edly less than is the corresrionning range of variation in the index of either sunshine or evapo- ration. This relation holds generally througho;:t the season. In short, the temperature efficiency values exhibit a smaller degree of geographical or Iceal variation than is exhibited by the index for sunshine and the evaporating power of the air .

Intrc'Iuctory. As has "been stated, the plant ineasure- rr,,--^-.•^B •*:i-'-i-^ —ar.o t'-i ":,-(-•■=>■! d'^i OH t t'vo -'V'f eT^s aftei' tiie T3fi '*■■,■■■ each culture irioliided stem height and leaf climensiciiG. 7rora these have been derived two 2-Vv'eek data in each case, which are g'vpr, in the t'^'hles, ^1) the relative rriean dallv rp.te of increase in stem height per plant and (2) the relative r.ean daily ra' e of increase in total leaf-product per plant, "both for the t'?/c-weelr ^-ioriod. As ■""^'''-ean --n-^ -oir"^'=^^. cnt, the mean daily rate ol. increase in total leaf-proiuct for a period of abo'.t 4 weeks is very nearly proportional to the corresponding rate of increape in actual leaf area, and it seems rafe to suppose as McLean did, that the 2 -week leaf-pro'^uot values are to he regarded as indices of increase in the area of t'- -. leaves. Therefore one of these 2-T7eek plant values represents the stem- producing po'ver of the plant ;and tlae other stands for its leaf- produGJ ng power, under the given set of external conditions acting duri-':'^- thp 2--,veek '-e-'-iod. Since the plants are talren to be alike at the start, fseeds) these two derived plant values should be the same for all individuals if all were subjected to t'^^ same eff'^'^'^ive pnv' rrrr.yr.fi-itsl TonSlt'nns throughout the period, and when the various plants are exposed to different environ- ments the values Jib t mentioned become criteria by which the ef- •f p -.*- i 'T ",;,-, Q g g Q-f f-,-op pnvi ro7-,ir,en t 'nay ^^ r>onr.,are^ with that of ail- other of course -.vith reference to the particular ^et O"^ internal conditions represented bj'- the plants at the beginning of the • tests. T.'ie tv/o -1 an-^ "-"^ =3 just me r. ^ ' '""i ^-^ ' i-'-'^y thus Tre regarde-l

5u

as i-elativ.- measures of the effectiveness or efficiency of the environmental coraplex fb r the 2-7,'eek period considered, as it acted to produce stem elongation and leaf-product increase, upon the soy-bean seeds employel in this investigatio:i. The values of these two plant indices are of ccurse given in the tables 'n terms of the corresponding average for all cultures considered, employed as.^unity, just as '^ +■'"'= -•-"■-' ^- -p the other relative valixes, and thej' all represent daily rates for tha 2"Week period in question. For convenience, the follov.'ing discussion will refer to the graphs rather than to the tables, but of GO":irse tables and graphs both present the same "data, in every case. This discussion will be given u:".der the three followin:~ headings: Correlations between the two plant graphs; Trends of the plant values and theit seasonal averages for the various stations; and Helations between plant and cl'matic gr;

Correlatinns between the t'^'o plant graphs. It is realily observed that the -^- c graphs showing relative rates '■-^' 'ncrease in stem height and in leaf-product have a pronounced tendency to exhibit the sar.e general direction of slope from perio' to

pei^iod, throughout the .season, for all stations. In many cases the two plant graphs not only slope in the same general direction fupward or downward) but their corresponding angles of slope are nearl^- "-'^e .^ar^'^ -■^:'' their corresponding ordinates are about . equal, so that the two graphs nearly coincide for considerable portions of their length. In other words, there appears to have been =i _ x ^^ :>-^ i^ced general agreement between the effectiV6x.-oo cf the surroundings to loroduce stem elonga-^ion and its effec- tiveness to increase the magni'^ude of the leaf-product, as shown by these cultures. If this agreement were per-*:--^ : "■

51

woulil iiiecni, ^1 Course, tuat the environment exerted "-' K>&;:ie i.xi- fliience upon the process of le af-siirfaoe increase fas measured by leaf-product), and either of these two criteria would "be a measure ■'-^' ^'-'^ ^ther. But ■^■" coincidence of the two graphs is not by any means -erfect and it becomes necessary to study their differences, especially with reference to the correspondinn- re- lative ^'■alues ""f -^-''^eir ordinates.

Inspection of the graphs shows that, leaving those fc r Oakland out of account the index of height increase is frequently greater than the other plant jnl"-- -^ \- '-•:- early and late por- tions of the frostless season, and that this relation generally is reversed 'or the middle -ortion. In other terms, the graph for stem elongation generally lies belcv; the other graph for the middle of the season and above it for the beginning and end of the season. In still other words, the seasonal maxima of leaf-product values are generally rela'^'^''^" y higher than those of stem elongation, v/hile the seasonal minima of the former are lower than those of the latter. It r;ay be stated, as an approxi- mation, , that vchen these f^-ro •■l?n+- values are both above 100 the leaf-product values are the higher of the tvro, while when both are below 100 the elongation values are the higher. In ^ho r.c,c--. .-.-^ Oakland, "^^--f^^ '^-Tilues ■''■-" -^ omparative? "' -"^-^ low throughoirt the season and y;hile the index of stem elongation reaches Bomewhat above 100 for two periods, this index is never surpassed in magnitude by the ' ^lo-'- ■■ ' _ "_--,,-,-.•: i-,-.-^ crease.

The generalization Just indicated, being a relation be- tween 'he rates of two -plant processes, seems ^c " --- a Physiologi- cal one, dependent upon the nature of the soy-bean plant, and

5<$

hence predetermined ty the internal conditions of the seed, Within the range cf enviroruaental conditions encountered in this study it appears th^-t the taller and more leafy the plant "becomes in the firs' ^ eek.o -. 1 growth, the lower is tlie v -lue of the ratio of final height to final foliar expanse. The two growth nrocesses here considered are therefore clearly inter- relate i «u that ilieither one alone is to be regarded as a cri- terion of plant growth in general. In the cases here s'^'udied it may be o f value to consider the average of these two indices as a tentative index of the general growth -^* ■^'-"^ -^lants during the first ty;o weeks from tiie seed, and inspection of the 2- werk graphs leads to the impression that averaging the two values is the most promising way to obtain :from them a single index of plant growth. The tv/o graphs are always so nearly parallel throughout the season at all stations here dealt with fnearly coinciding for many periods, ai las been stated) that the charts have not been further complicated by intro 'ucing these average graphs, but their form is readily appreciated from the two graphs that t.re given. Still another possible way to obtain a single index for the growth-behavior of the plant as a whole may be obtained by determining the ratio of the 2-week rate of stem elor.-' "^^ "^a to the corresponding rate of Itict.' -ijru.iu.it in- crease.

The general relation between the two plant values, upon which t: e discussion just given ia based, does not alwa- s hold, however, and tiie following more detailed discussion of the plant graph forms for .individual stations will be o-f' v;lue in showing the main exceptions. As has been mentioned, for Oakland the height value lies above that eaf-product throughout

5:i

the season. The tvyo graphs have the sarae general direction of slope excepting between the periods beginning July 15 and July 31. For Ghews"ille the height graph exiaibits the sarie f?en-. eral direction of slope as does the leaf-product graph from period to period, throughout the season, and the former lies v,'ell above the latter for the last 4 periods (beginning Aug. 25, 3ept.ti, iSept. S2 and Oct. 7j. r'cr the periods beginning June 16 and June 30 the aame relation holds, although the index values are large, especially in the case of ths first of these tv/o periods. For Lonrovia the two plant graphs follow each other very closely throughout the entire season. The graphs for College, for the periods begirjiing July 13, July 17 and July 31 illustrate ^he teniency c "'^ the height value ' decrease rela- tively to those of leaf-product 7:h.en both values are large. 7or the periods beginning Sept. 10, Sept. 25 and Got. 10 fOr this station, no.v ver, both v;-.' _ \r e small and still the height lies above the other, and for the period beginning June 19 the relation between the two graphs is reversed, al- though both values are relatively high. For Baltimore, the periods beginning June 10, July 23 and Aug. 20 exhibit exceptions to the proposition thiit the height graph should lie below the graph of le af -product when both plant values nxi^ Large. The generalization is true, however, for the period beginning Aug. 6, in which case the values are both lars-e anl the leaf^^ product graph lies well above the other, i'or ijariington the two graphs agree very v/ell in form throughout the season, ex- cept for the period beglnnirig July 10 in which case' the gen- eralization holds and, .vi th both plait values high, the one for height is considerably lower than t' : ar. For 3cleman the

54

generalization holds faix"ly wall. Tor Easton the gereraliza- tion hc7 l3, v/ith. three e:'MeptionP: the height j.adex is lovrer than the other for the period "beginning May 8, "Itihough both' indices have low val'aes, and th,is relation is reversed for the periods "beginning June 22 and July 20, in spite of the fact that both values are Ir^rrp-R^r. thenn o-ses. For P:M?-!nes£ A^ine the gr'.-'phs show the values ol' ti.e h6isfi.t index at nigher than . those of the other index for the periods ^sginning June 23, July 7 and Jvlj £1, although both jndir;es are large for all three periods. Otherv/ise these graphs agree with .the gener- alization.

The fact that the genflnl i p^a'^i 'r, gl-ron sir;'--- '-.olrls i r the great majority of the cases here studied renders tl^e exceptions od special interest vlth reference to the causal relations In- volved. ■ I": maybe supposed, ssuming that the olants v/ere all ali'-e at the begirming of all cultures and that no disturbing influence was introduced by soil conditions, tin t -the periods characterised by exceptions to this generalization should also be characterized bi- some ;3ort of corresponding peculiarities in the aerial environmental complexes. No^.-, ^ stu:Iy of the

hharts for thii= exposed s■f'■^ *' fr-.Q >rings 'n-^ *"''"e follo:-;ing fact; most of the £-v;eek periods for «iiich both plant values -are large and yet the index of stem elongation is greater than tha^ leaf-proiuct Incre'se, are characterized by " "^ 'ices of oil"- shine intensity. Thj.? suggests that the plants ^f these cultures experienced an acceleration in *:' ir rates of stem elongation due to low ligi't 'ntensJt.y, iv + *-Hat they exhibite"! sr-.^ r ''

the effects of inci^jient etiolation, Hhej see 3hcw a some-

what increased, rate of ^ elongation and a somewhat decreased

55

rate of leaf expansior, a -••-.r'-D v-^ -: •-; -Kr rl->y+:j -rnnat y' -nc lo-fp

light radiation. This interpretation is not to be regarded as at all well established, but it is at least a suggestion of one •way in \vhiGh the external 3onri i -^ ^ "->. rf light ' i- -^ « --i ci • ^ .r ;vn ' duration may be registered in such plants as were :ie re employed,

Trends of the plant v lues and their se.i^.gonal ranges for the varioas stat'ons.

""'■e fo!lk)v-ing o nsifler--) ti nn r-f-' th --^ seasonal marches of the i:-Y.'eek plant values ior the various stations will be limit- ed in extent, since most of the facts and deductions that seem to be of i "-nortivir;.-! • r, •*-his connection can be befitpr ''r,■rr^^.^'f out later. Attention v;ill here be called only to two generali- zations regarding the plant grap' s: (1) The plant graphs be- :'■ - -.-if-.', "alues of e.'^or.f. I'^n, r'se to hif^h midsurr'"^ f-"'" -'alues and then fall to low values .it tiie end of the season, (2) Differ- ences in the magnitudes of the midsummer maxima constitute the chief discre'na:'!''^ i ;= ^> ^, e'*"'vfle'''' ^'^^^ graphs for the -/srions ^f''''ions,

Oakland siiows lower values of the stem height and leaf- product than any other station owing largely to the low values §f thfl ■f"err::^eratT)re th. ■•■ -r, r^v*. i1 .-^--l .-'■♦- fhis of --^ i ,-,r ^.'■irnr. p-hn.-i*:

the season. The Jata ' ' 's stu.ly indicate in all cases f the climate of Oak:" o far as it affects the plants

lite that of B.'ny o"' thp "jth.^r sta"*"' '>'.;-■ p ir-r,l o-7p i . ■R'-'-'- graphs for this station, show the, typical low values at the be- ginning and end of the season, however, with midsummer maxima of 124 for the stem .x^.^^,. ,.....,. ^^ :or the lecr..I-..r.vi act . Ji.t, - "ille shows tj^pical graphs, the highest value reached by the stem hei-ht being _.14P) and by the leaf-product 139. The end of

'6 S e u .-i V.' Xi ctt this bL J liiix ci'J tuij i; ►j'j. ij ■■; „ .,

5 b

Oi it!:ii-pi-'- *" . T'xit' ^x'uiJ^-b I ur Llonrovia are ^iao typxo:.±, with low values for the period ■beginning May 18 and low values at the end of the season after a nildsvunmer m.a::timum of l,ci '-'.'V stem height ' l-""" 'V le-j.i-j..>r<.auot . An explciiiation for the low values shown by the periods "beginning Sept. 7 and Oct. 8 may lie in ^he fact that the ninimum temperature dropped in both these ^j«ix'.. ih tu u. point only severax i«grees tbove freezing. IPor both Ghev/sville and Monrovia the plant values are for the most part less than 100, ?;ith rela"^jvely lo-.^^ niidsuriner claxima of 132 and 203 for the stem heigh" itui ±~u'' product, re- spectively. The Baltimore plant graphs b = gin -vith high values, and reach maxima of 183 and 233 for stem height and leaf- product respec^^ively. For Darlington the main feajtures of the plant graphs that serve to distingaish this station from others are the very high maxima of 226 for the stem hejght and 295 for

leaf-product, for the period beginning July 10 :nd the x-rxative- ly high values shov/n by the graphs for the beginning of the season. The midsummer m.axiraa for Ooleman are 157 and 211 stem heir'^--^ ■ '■ leaf-product/, respectively. The plant graphs iv:- Easton show relatively low values.of the midsummer maxima for the plan"*" indices, 14G be'ng the highest value reached for height ciiiu. x6-5 iui l«a.i ~i;i 0 i uu I , Also, the plant va,l..«.sd are lo,- ' ■'■• the beginning of the season for this station. The midsummer maximum for the hemght value for Princess Anne is 197 and the correspon d- Ing -naximuni •'"■r ^he leal-px-uau-ot value '" l'*3.

T':e olant graphs as may be seen from the above statements of their main features , fall into three groups: fl) The Oaklai^d , graph s . These values of the leaf-product index

below 100 for all periods and' similar low values of the stem-

57

height index foi- all pei". except those beginr. ' and Aug. 14. "■ ' ^'lese graphs are not rela-^ively hiigh. ;'l) The grapiis ior JJie-.vg ' , ' onrovia^Princess Anne and Sastcu . liigher raidsumn^r valu ; . lant growth rates than, io the Oakland graphs, the maxima "being about' 1 l/2 times the seasonal average here Gonsiderei as 100. (Z] T:ie grapi.s for Coligge, BaltiiTiore, Darlington, and Ooleman -re peculiar in tha-^ they exhibit high relative values of t'reir maxjm^i. From the present poin"^ ' -le'w then, the ; nterestii-g character- istics of the graphs are summarized ' "^he above classifjc- tion and the statement tftt thev show a seasonal march from. Icr: values in the early part of the season to high mia.sur.".;.ier values Tith a subsequent falling off to low values in the later parts of the season.

Correlation of plant and climatic da''"a.

In attempting to Gorrela-f-e the plant and climatic data the graphe representing the plant and climatic m.easurements for all the cultures of the Investigation were inspected to deter- mine -^^hether a general scheme of correlation could be ■'''•:"mu- lated that would explain in any consistent manner the cr-aructer- i sties of the plant behavior in terms of the corresponding cli- matic values. It was fo-;nd po;--.sible t? formi:l?,te such a scheme but sir.ce each of the three climatic conditions measure a acts on the plant in a ni:inber cf ways, the most that could be accomplished wo.s £. ieterminati on of what seemed -'■he ■pre'^on^.ero.t- ' ng influence on the plant of each of *" t'lree envj.roi-r.ei^xa-i- condif ':.ere considered. conclusions reached as to the principal effect of these environmental conditions are given

5H

"belo':.-. Subsequent re^'^erence to the i'lta v.n ■oresented in the gratv.B v'ill aiiOY. the extent to v.'hicii u:.e iGiie:ne adopted .':ia;." be applied. T/ ^relation scheme is not to be oonsiderei .as final

in any sense, but '^^erely -as an assunption upor. h 3. fairly

consistent tij^ulcinci* ioji, <.j i the data of the present study ii,a,/ be :nade .

The most reasonable basis for correlating the plant and climatic graphs^ t-e determined from inspection aeeuio to be oiLe that assumes growth as conditioned mainly by the temperature, sunshine having a direct effect, and both conditio-'s acele'rat- ing growt'- -^-cesses as they increase. Evaporat ' "- -;

assximed to have an 'nverse effect on growth, the latter decreas- ing as the former increases. These assumntions may be stated ' :■: the follovn.ng .;.a,iit.»^r :

f fT)x i(L]

r-

f (1,)

in which r represents the rate of the plant process under consideration, T, L and ^ represent temperature, I'c-'^t, and evaporation resr^ective"^ - ". .'3ome sucii simple workix.g li^'putlietl .3 as the above facilitates the attacking of the complioated prob- lem presented by a group of nlant and climatic data such as ib l.«-.;e pj e;^t;r.';ed . ,v« a'. "^ , '' '.^iirsfe, know ?:hat form the fumtions of T_, L, md S will have, but it is probable, at least, that _T and L are in the numerator cf the fraction whi'^h is her^^ uaei "" - .,.,-,^.- ,„ ,^ ^..-x.,^^y^Q-._~- -^'-n rela'^' " between tiit; pia;:ts and

the climate, :ind that H is in the denominator.

The -olant rrvi: >-^-n PTa"'"ii 'ip "i , "-prioiT by ner" nil, -"cr t'le

various stc^ions \7ith the above apsinapt ions in mind might be

5IJ

expected to ahov.- certain relations to the cliiaatiG vcLxaes. If the asswoed relation holds it would be supposed, that the plants woiild show a high growth ra-*"e Thenever the temperature value is high unless at the sane tir.-: " ' ' -r light iu v'ery fi.e. shovinp- a value of 80 or less) or the evaporation very hisrhfi.e. shoTTing: a value of IS 5 or .Tiore), V/hen the temnera- turs index is Ic^^/, the plant gro\7t!:i would be expected to o'lov/ low values, the amount by v/hioh it is depressed being less in *"hose r^eriods for which -^he other climatic conditions are favorable .than in the periods fur 7;:-ica they are unf avoi-able. "High" and "low" vrill be used, in general, as above and belov: ■'rh''' seasonal avera^re for the sta''"e, since it is obviourl^' im- juLbj/ole ill :::.■■: i^uali tative method of treatment used in -^his study to assign any definite values to these terms. Also the correlation betv;een the riant and climatic data for the two-week periods will take into c..o';jount only the leaf-product. This is done for the reason that the leaf-product, ie approximately proportional to the leaf area anl dry weight for soy-beans

V Hildebrandt, ^. ". , Leaf-product as an t nder of growth in soy-bean. Johns Hopkins T^niv. ']irc. Inarch, 1917.

as has been sho-Ti by Mclean md by t'-.e wr'ter in a r;revioiis paper; and ,:it ^r^ 7.yj.gnt of plan's is ^as most generally accepted criterion of their growth. The previous discussion of the relation bet^reen leaf-product and ? tem height, has shown tha-^ they - ■ ■ ' • - .'• -_ ■ .. ^ aultur-^^^,

to the same extent. A considerr" ' of the stem height in

80

this part of the paper would thus involve repetition and 1p

unxiecessary.

?or Oakland the letii-pru ".act graph follov.'s the ^ . -

ture graph from the pe riod beginning June 5 to the peri od be- ginning July 31. Fcr the '^reiods July 31, Aug. 14, ani Aug. 27 the te;;:ptiritture remains i-.ppro.-iuiately .y^i.titant and '-'- ' ^- - '' leaf expansion follows the light graph. Light parallels tempe- rature, and both parallel leaf-product from the -oericrl beginning July 19 to the period begir-ning July 31. The questic: ' ■ ^' which of the two factors is the determining one here -.'culd seem to be answered by the be' avior nf the plants '''or ■'■'^e r-ericds beginning June c, J^ily 31, :rag. 14 and Aug. a. in the oeri^..i beginning June 5 the plant graph descends wilfch the temperature efficiency valae although the sunshine gratih rise-^. ail ir, -^he periods be^-inning July 51, Aug. 14 ana Aug. .:/ 7/nere tne teir.- perature remains nearly constant, the leaf -product fcllov.-s sunshine intensity. The v^ry high evaporation ra-^e for thn period beginning June 5 seems net to ''ave had much effect on th«= plants. loT Chewsville, leaf-product follows the temperature index graph during the greater part of the season. ''^he pla^it gtaph shows. a tendency to rise &' " the graph of temperature values during the first part of the season and to dx'op below it at the end, this being- due nerhaps to high values of sunshine intensity luring the eiiriy ".ericas and lo^r values ilurj.ng the later ones. For the periods beginning S^ept. 2£^ and Cot. 'i leaf-product follows '■he r^rarjh of light intensity, the tempe- rature value remair' e.rly constant. It vd 11 be noted th t the fe riod beginning Aug. 11 with climatic conditions ap -arently fqvcr.nble for -rowtl- aVi ---.rs' a low leaf-nrcduct . "cne of the

61

evaporation valii'^- '''^'■^ +-n ; o stat''^''' ■■■>'=e'~'_^ ^■■' '- vp v-^r 'r\rc\- e'nough to affect the plaiits. Leai-px-od-uct at Monrovia, follows the graph of temperat fare index values in general. In the r,eriods

■j,g^; -,,„ ;^,^T:,v,o 1^ T,v':o !-;_ ii-a-T 13, and -Tiily ?7 !-'p-! .;

evaporation probably depress the rate of ^e af expansion since the temperautre Indices for these periods are well above the seasonal av'^^--'^ for the st-^te, ' ^O'^ ^.-p-.p reason ^^^-^ apparent from the climatic graphs, the period beginning Aug. 10 shoves a relatively low value of leaf--orodiict and the perJ ,":inning

Sept. 21 a rel ■•^''' ' ■^^l '' h' rrh -"^^lue. T^i-.p -.-\ir^-r\^ ^r^^'" ^ .-\-i- icl"''™? shows a number of peculiarities -vhich cannot be a^dequately cor- related T:ith the climatic d&ta. Leaf-product has a relatively lower val"^ ^'•^'^ pp -•,-■! beginning June 19 and a r^i ''■■ ^i '"■'^>'

higher value for the oeriod beginning July 17 than -.voiild b-: ea.- pected. The plant graph lies well below the temperature graph from ■'■h^ ~"riod beginning Aug. 14 to the ^n"! of the "-epscr; due probably to low values of sunshine intensity. 'Ilhe light data are not available for this station, however. The rise in rate "--* ] eaf-produc-^^ ':-■■' ^rnv the period beginning IJay 23 to the period beginning. June 6 may be explained by the corresponding fall in evaporation rate, althoiigh this fall is sljght. Baltimore shows relatively hi-^i. vi;.:.u.ey uf leaf-product for the periods beginning June 25 and July 9 as would be ex acted from the fact th-t the temr;erature value is hi^Th and the evaporation lov,' for these eerioas. one ■: ' ■ -oroau ' .e

period beginning Aug. 6 is not clear. Darling'-on is :Ti stinguish- el from th-^ rest of t: " ' ' ", very high leaf-product for

the. period beginning July V "'-^ ralue l being 295,

6S!

practically;- three times ^'^•^^ ■ >-'■■ ^■"■r ^'^e state. It -ill he

observed that this period has high values of sxmshine and of the physiological temperature index and a Io-a' value of evapora- tion, s, ^combination o-?' r^n ■■-,'\ ' ■^.: r.-- c. -•■^•-.'o — r^iild be expe'?t'-'' *<- '^p- optimal for plant grov/th. Tiie iiigii value of leaf-product for the oeriod beginning -June 13 as compared to those beginning May ?n pr^"! ra''' ^"^ i" ■'^'••^"cbably related to the r-"; m*: ' v^iy low

evaporation rate luring the ■)eriod beginning J'une_ 13, The

sWovj Similar- o<vvvvd*t\"a-',aVfi^a£<e^ bt^T"ti^4, pJa^rr c'.ccrSi periods beginning 'July 10 and Aug. 7 for f-e latter are lacking

and no r>c- t \--.r * p^^r^ o-^* ^hp rerio-lK • :^ !'p,P8ible. ""or Col^inan, the leaf-pi-oduot follov/s the graph of temioerature values in gen- eral from the period beginning May 13 to .the period beginning July 22. The relafivoly lovv values of t^'- l^--^"^"-n-,-oduo t- for the periods beginning Aug. 5 and Aug. 19 cannot be Scv'-isfactorily correlated with the climatic dati. For Saston, high values of the temper -ture indices seem to account for th^^ iii gh values of tibe leaf-product for the periods beginning Jtily 6, July 20,

r

Aug. 3, and Aug. 17. The graph^leaf-product lies well below the grai-^h of •^er'-»-^'^"!"«+^nr'= values dur J - - ^'-^^e last part of t^"^ T!be period beginning Aug. 17 would be oxoeoted to ehow a higher value for th'3 plant grovth r.i, te than is actually found but the lOY/ value of simshine for this period --- explain the xu, plant value. For Princess Anne the leaf-product graph follows the temperature grar)h practically throughout the season. Evapora- tion and sunshine intensity are both lor,- '' - ^'-\s stat ' evaporation would therefore not be expected to influence the plants. '^he low sim?.h5r aes for the periods beginning

June 23, July 7, .;,.:„, .x, .. ,- ' .-. ay be related to

the fact thaJt leaf-product aues not show values as high for

63

these periods as would be expected from the behavior of the plants at the other stations.

Before concluding this part of the discussion it is de- sired to call attention to some peculiarities of the plant graphs which cannot be correlated with the climatic data, "^he first of these 's the occurrence of growth rates for some of the oeriods very much higher than tJie climatic data and the assumptions male above would lead us to expect. This condi- tion of afiairs in illustrated by the period beginning July 17 for College , the period beginning August 20 for Baltimore, the one beginning July IC for Darlington, the ones beginning June 24, July 8 and July 22 for Coleman, and the periods be- ginning July V and Aug. 4 for Princess Anne. In all of these cases climatic conditions favorable for growth would seem to ob- tain. The climatic values for these periods do not, however, appear to be sufficiently different from other periods showing relatively lower plant values to account for the very high plant growth rates. This behavior of the plants may be ex- plained by the fact that two periods showing the same average intensity of climatic condtions may differ greatly in their plant producing power m account of a different distribution of the high and low values of the intensities of the conditions luring the period.

In the discussion of the two-week temperature data at- tention was called to the fact that the graph of temperature Values shows two maxima, and that this was corinected with a peculiarity in the behavior of the plants. Th° peculiarity ther'' referred to is that the olants do not, except in the case of the culture at Oakland, respond to the second +-emperature

64

■;ga'luoe-iB.--th^ maximum by a corresponding increase of growth. For C!hewsville» as an example, in the period beginning Aug. 11 we have a low value for the leaf-product with a high tempera- ture index and the other conditions at about the seasonal av- erage. For Monrovia in the period beginning Aug. 10 with high temperature value and sunshine at about the seasonal av- erage the plants show, nevertheless, a relatively low v ,lue of the leaf-product. This may be contrasted with the period beginning June 15 at this station which, with a leaf-product about the s^me as that of the firs^'-m'^ntionei period shows less favorable growing conditions, namely, a much lower rela- tive temperature index, a very high evaporation rate and a sun- shine value only a little higher than the corresponding value of the period beginning Au^. 10. For College, the oeriods beginning July 17, July 31 and Aug. 14 with about the same values of temperature and evaporation show magnitudes of 152, 203 and 150, respectively, for the leaf -product. This varia- tion may oossibly be related to differences in the value of sunshine intensity for these periods, but the sunshine data are lacking for this station. For Baltim:re, the periods beginning July 23, Au-j. 6 and Aug. 20 show large variations in the leaf-product with comparatively slight variations in the climatic conditions. Evaporation is slightly less in the per- iod beginning July 23 than in the period beginning Aug. 6 and considerably less in the period beginming Aug. 20, but this seems to be without the expected effect on the r)lants. For Ocleman the plant graph siopes upw.rd to a value o-^ ovr 200 for the period begirjiing July 8 vrhile for the oericd beginning Aug. 5, which has climatic corditior.s apparently as favorable,

85-

the relative value of the leaf-product is 138. l^'or Easton the leaf-product is lower than would be expected for the peridd beginning Aug. 17 and for Princess Anne, the nlant values for the period beginning Aug. 8 are much lower than for the oeriod beginning July 7 which had approximately the ssme climatic conditions as the frTst inentioned oeriod.

A third peculiarity of the graphs that cannot be corre- lated with the climatic data is that the stem height reaches its highest value for the season before the leaf-product for all station except Sarlington and loleman. "^or "Darlington, the highest value for stem height and leaf-product both come in the period beginning July 10 and for Coleman the maximum value for stem height comes in the period beginning July 22 while the leaf-prodcut reaches its highest value for the season at this station in the preceeding period. At the re- maining sta-^ions, the highest value of stem height occurs two weeks or a month earlier than the highest value of leaf-pro- duct.

The foregoing discussion of the two-week data empha- sizes the obvious fact that the problem of corjslating the plant and climatic measurements is an exceedingly complicated one. Assuming that the conditions of the experiment give approxi- mately constant root environment, the plant growth rates are a function not only of the climatic factors, three of which are measured for these studies, but of the conditions within the plant as well, TJ'e do not a^^ present know how to measure these internal conditions. The g^rowth rate is thus a function of a number of variables some of which are known and some un- known. The object of the preceeding discussion has been sim-

86

ply to emphasize such relations existing within the .orroup of climatic factors, between the two plant measurements, and be- tween these two groups of data as can be seen by inspecting the graphs.

67

The four-week plant and Gllmatlo data. Exposed Stations.

The general plan of treatment adopted in the case of the two-w^ data will be follwed in the present Usoussion of the four-we°k periods. It will Le possible, however, to treat the four-week climatic indices briefly since the graphs of these factors for the longer growing periods cover approximately the s ime season at each station as the two-we-k climatic graphs. In accordance, th^n, with the plan previously laid down for dis- cussing the two-week data, three divisions of the four-wet^k data will be made; (1) A brief consideration of the climatic graphs; (2) A consideration of tho relations between the plant graphs. f3) Attempts to correlate the ^lant and climatic graphs.

The four-week climatic data.

It will be rememHiered that the cul turns were started every two weeks and that each grew for a nericd of four weeks. The four-week periods thus over-lap, and, as has been mentioned, avera?-es of the climatic factors for these overlapping periods fcrm a smoother graph than averages for the two-werk periods. The result of averaging over the longer overlapping periods is to eliminate from the graph all the smaller fluctuations. The fTir-week graphs therefore show the general seasonal march of the index values for various stations better than do the two-week while the latter show the details of the seasonal

68

mar^h better than the four-v.-efek graphs. Thip fact will be brought out by a brief reference to them at this point.

The values of the temperature indices for the four-week periods show the seasonal mar-^hes of this condition for the various stations, from low values in May to high midsumner values and then to lov: vmlues again in the last part of the season. The grapn.s for all of the stations except Oakland show a steeper slope after the midsummer maximum has been passed than for the periods during which the temperature was rising to this maximum. The two maxima which were present in most of the two-week graphs ar^ eliminated in the four- week averages and the graphs of temperature values show in- stead a period of about six weeks during which this condi- tion remains approximately constant.

The four-week evaporation and light data, show the gener- al characteristics of the seasonal marches of these -conditions previously noted as exhibited by the two-week data. It will be seen in the first place, that both graphs exhibit a downward slope from the beginning to the enJ of the season; and, in the second place, that both graphs show, in addition to their high primary maximum in the early part of the season, one or more secondary maxima later. In some ciases the secondary maxima in the evaporation graphs coincide with temperature maxima. Both of these general characteristics shorrn in the four-week graphs of evaporation and light are shown by the two-week graphs but since small variations are eliminated by averaging the overlapping periods, there are fewer secondary maxima in the four-week graphs. In the case of evaporation, there is usually one secondary maximum occurring in or near

6:j

the four- week period including the last two we^ks of July and the first two weeks of August, In the case of all stations, this is one of the three four-week periods showing high tem- perature values. The foiir-week climatic graphs, -^ill not be taken up further here, ''^he method by which the four-week data are derived from the two-week data amounts to the same thing as smoothing the two-week graphs and only the more pro- no'inced characteristics of the graphs remain after averaging. The interest of the four-week climatic data thus lies mainly in its relation to the plant growth rates.

The four-week plant data.

Relations between the olant measurements.

Certain general relations were pointed out, in the case of the two-week data, between the stem height and the leaf- product, and mention was also made of the fact th .t the leaf- product numbers showed only slight differences in value from the leaf area and dry weight numbers when the actual growth rates are expressed as in the present study, that is, using the average of all of them as a unit to which to refer the Individual rates. In the four week data, the rate of stem elongation may be compared with the rate of leaf expansion as determined from actual leaf area, instead of with the leaf-product which is used in the discussion of the two-week data as an index of area, '^he comparison between leaf area and stem height for the four-week growth periods shov/s the same general relations as appeared to exist betwe;~n stem-height and leaf-product for the two-week growth periods. Owin? to the fact ihhat the plants

7U

were grown for a longer time, however, there are fewer cases in the four week periods where the rate of stem elongation is greater than the rate of leaf expansion. In most cases the rate of stem elongration is well below the rate of leaf ex- pansion* This relation again illustrates the teniency of the soy-beans to show a low rate of height growth relative to the rate of leaf expansion vrhen toth rates are large. A conside- ration of the riant graphs with the purpose of bringing out the relation between stem elongation rate and the rate of leaf expansion follows below.

For Oakland, the stem height g raph is above the leaf area graph for the first three periods of the season and be- low it for the other five neriods. For ^hewsville, the three plant graphs follow each other very closely anci. the differences in their relative positions are probably due, for the most part, to individual variations in the plants of the separate cul- tures. The Monrovia grap s also correspond closely with the (e^fteep t ion • -e-f- -tfe^—fao-^ that stem height shows a well-defined tendency to remain below leaf area during the first part of the season. For College, the stem height is ";ell below the leaf area for the entire season e: cept for the two periods beginning June 19 and September 25.' The Baltimore graphs show stem height values higher than the corresponding leaf area values in the periods beginning May 14, May 29, June 10, and Aug. 20 dae to low light intensities as shown by the graph of sxinshine intensity. The Darlington cultures show very high values of the plant growth rates with stem height below leaf area for the entire season. For Golsman, the stem height graph remains below the leaf area graph for all the periods ex-

71

cept the last where it rises very slightly above the leaf area value. For Easton, the -nlant growth rates show nearly the ssme relative values for all the Gul-*ure periods of the season. For Princjess Anne ^he stem height and leaf area graphs shov/ a de- parture from the usual behavior during the first three per- iods of the season. E'er these periods, leaf area is relatively large and stem height relatively low for some reason not appar- ent from the climatic graphs.

The most striking ralation between the plant measu^^e- ments for four-week periods exists between the leaf area and the dry weight of the plants. It will be seen that for most of the cultures these two plant growth rates have practically the same relative numerical value for any given period. The Oakland graphs show this in all periods except the period be- ginning June 5 where dry weip-ht is well above leaf area. For Ghewsville, the relation shows very well throughout the sea- son. For Monrovia, the leaf area number shows a rather large deviation from the dry weight number in the peri ods beginning June 16, June 30 and Aug. 25 but otherwise ^he two growth ra^es correspond in relative value during the antire season. The College graphs show close agreement, with dry weight above leaf area during the firs'*: part of the season. At Baltimore, the relative leaf area value differs considerably from the rela - tive dry weight value in the periods beginning Aug. 6 and Oct. 1 but the remaining periods bhow close agreement. The Darlirigton graphs show close agreement for all periods, "^or Coleman dry weight and lea-^ area agree well for all periods, except those beginning Aug. 5 and Aug. 19, and for Easton no large differences occur for any of the cultures. For Princess

7Z

Anne, the period beginning Aug. 4 is the only one v<5howing a difference of considerable magnitude between the relative leaf area values and dry weirht values.

The property of soy-bean shown by these graphs renders possible the use of the leaf area of the plant as an index of its dry weight. The bearing of this property of soy-bean on its use as a standard plant for climatic investigations has been referred to in a previous paper by the v;riter.

f

Hildebrandt, y. I.T. , Leaf-product as an in'lex of growth in soy-bean. Johns !Ioi-i>ins Univ. :;irc. March, 1917,

The correlation of the plant and climatic values for the four-week periods will be discussed under two heads. The first of these will take up the relation of the four week relative stem height values to sunshine and temperature iridex val^^eB and in the second an attempt v/ill be made to correlate dry weight with the tJiree olim-.tic conditions using the assumptions made in correlating the two week leaf-product values with these con- ditions. The dry v/eight is employed in the consideration of the four-week data ra'-her than the leaf area since it is the most frequently used criterion of growth. Also, as has been noted above, the leaf area and the dry v/eight of the plants in this experiment correspond closely in relative value, ind for such rough comp-.risons as are here made, may be used inter- changeably.

•>

7;^

'Correlation between nlant and nlimatic data.

Ill ^ —

Relation of ptem heij^ht to sunshine ani temperi:!.ture . No very definite correlation bet-een the stem height and climatic conditions has been found. After the plants have been grown four weeks, ho^'-ever, the graph representing the rate of growth in height bears a general resemblance to the graph of physiologi- cal tempr^rature indices. In the first periods of the growing season, also, ■*:he data indicates that the s-^em height is in- versely proportional to the sunshine intensity, this relation being indicated by the fact that the graphs of these two quan- tities slope in opposite directiO' s in many oases. The effect of the light is, however, only secondary and does not disturb the relation apparently exinsting between stem height and temperature to any considerable degree. These two relations will be brought out by an examination given below of the graphs for the individual station.

For Oakland, the stem height graph and the graph of tem- perature values show the same -^eneral form, and the stem heirht and sunshine graphs are opposite in the direction of their slope from the nerlod beginning May 23 to the period beginning July 31. Fot this station, tae other plant graphs parallel the height graph approximately and thus no conclusion could be drawn from the data for Oakland as to whether the difference in slope direction of light and stem height is accidental or due to a real effect of sunshine intensity on the rate of e3)on- gation of the plants. A t some of the other stations the height graph shov;s an opposite iirection of slope to the graph

74

of sxinshine intensity, however, when the other plant graphs do not parallel height as they do approximately in the present instance. For Chewsville, the graph of temperature values and the height graph have the same general form if we neg- lect the period beginning Tune 16, but here again he plant graphs all shov/ ordinate values so nearly the same th it no con- clusion c-n be drawn as to whether the hei|ht of the plants 3s affected by the climate in the manner noted above. The Monrovia graphs, however, support the assumption made as to the way in which light and temperature values are related to stem growth. It will be observed that at this station the general form of the height graph resembles the general form of the graph of temperature indices. Prom the period beginning May 18 to the period beginning July 27, the graph of stem height slopes in a direction opposite to the graph of light intensity. The height in these o\iltures is apparently responding to the cli- matic comolex more or le.is independently of the other plant measurements as will be seen from a comparison of the three plant graohs. The graph of stem height for College has the same p-eneral form as the graph of temperature values and is well below it for all the periods except the last. For Baltimore the graph of stem height would indicate again tha""- this plant growth rate is responding somewhat independently of the other two. The graph has the same general fotrm as the graph of temperature indices and shows an opposite slope di- rection to that of the light graph during the entire season except between the periods beginning June 10 and June o5. and between *-he periods beginning Sept. 19 and Oct. 1. The plant graphs for Darlington all reach abnormally high values. The

75

resemblanee in general form betv^een the stem height and tem- perature graphs for Darlington is thus less striking than for the other stations. Tne effect of light intensity on the stem height for Darlington does not show in the graph. For Coleman, the general resemblance in form bet-^;een the stem height and tem- perature graphs is quite close. The light and stem height graphs slope in opposite directions for the interval between the period beginning May 13 and the period beginning June 11 and the interval between the period beginning June 24 and the period beginning July 22. The stem height graph for 'Gaston shows a close resemblance in general form to the graph of physiological indices nd is opposite in slope to the sunshine intensity graph from the period beginning May 8 to the period beginning July 20. The stem height graph for ^rincess Anne shows the tjrpical reittioh to the temperatu'-e index values and is opposite in direction of slope to the sunshine graph from the period beginning May 11 to the period beginning June 23 and from the period beginning July 21 to the period beginning Aug. 18.

There are certain obviuus objections which may be nade to the above vie- of the relation betv.'een stem height, tem- perature, and litp-ht. It may be said that the opposite slope of the stem height and sunlight intensity graphs during the first part of the season is pi.rel" accidental. The stem height, being apparently determined in 'he main by tempera- ture values follows the graph of temperature indices, which graph-iuring the first part of the season has a direction of slope opposite, in general, to the direction of slope of the light graph. Attention has been called, however, to a number

76

of cases, of whioh the Monrovia data are a good example, in which the light appears to clay a definite, though secondary part. That it is playing such a part is indicated by the fact that the stem height graph, while corresponding to the tem- perature graph in general form, shows minor variations in slope noted in the course of the discussion which suggest an inverse relation between stem height ind sunshine intensity. Another objection to the interpretation lies in the fact th t at the end of the season ..hen sunshine intensity is very low the rate of stem elongation falls off very rapidly. Thip is not what we would expect on he assumption that li ght and stem elongation are inversely proportional. In ans^^/er to this, it may be said that evidence can be secured from the data tending to show that the low light values at the end of the season depress the rate of photosynthesis in the plants. This evi- dence will be giv n in a later paper. If the rate of photo- synthesis is lowered it is reasonable to suppose that there may be a defJ.ciency in the amount of nutrient available for growing regions, resulting in a diminution of the rate at which the various growth proce sses take place, T?e -^hus se- cure a possible explanation for the relatively great falling 0 f in height, ( and in the other growth ra'^es, as well^ occurr- ing at the end of the season. A third objection to the above interpretation is that the inverse relation noted between sun- shine and stem growth exists also between stem growth :.nd evapo- ration. The question as to whether this relation ia an acci- dental one due to the correspondence between the r:easonal m rches of sunshine and evaporation or whether evaporation and not sunshine is the determining factor here, is rather definitely

77

answered by the behavior of the soy-beans 7.hen ^rown under glass und in a forest. This behavior indicates that evano- ration has only a slight effect on stem elongation as com- pared to sunshine intensity. The consideration of the covered and forest station which will appear later, will justify this statement .

Relation between dry ?:eip-ht and the three climatic con- diti ons. '^he last part of the discussion of the four-week

data will attempt to sho^r that the dry weight of the plants and the climatic conditions may be correlated using the assump- tions nade in correlating two-week leaf -product and climate. As was no'ed in the treatment of the two-week d tta, these as- sumptions may be expressed in the following way:

fTL) X fJ^Tj

^ ~ ""tlij

in which the symbols have the same meaning as in the discussion of the two-week data. It was previously brought out that the above equation states the rate of growth to be directly pro- portional to some function of the light, directly proportional to some function of the temperature, an' inversely proportional to some function of the evaporation, f (L) , f[T), and -ffEj are used in the equation instead of L, T, and E since the environ- mental conditions do not affect the plant in a simple direct way, but bear a complicated tmknown relation to the growth rate. The physiological index, used as a means of expre ";sing temperature in "rhis study Is an attefipt to evaluate f(_T) directly- Ai has b en suggested by Livin -ston, corresponding indices

78

might be secured lor tie direct evaluation of f (L) and f (5) . At present, however, such indices are not scs.-.=ye^ available, and relative daily rates of evap ration and relative daily sunshine intensity values may be used to represent f(E) and j^fL) .

Tn correlating the four-week Glimatic and plant graphs it has been found, as in the oase of the two-week graphs, that the behavior f certain cultures aannot be accounted for by the climatic averages for the period. In some of these cases, the fo r-week plant values seem to be determined largely by the conditions during the last half of the cultiire period. This seems especially to be true of those cultures which made a relativ'ly small growth diiring the first two weeks from the seed, as occurred in the case of most of the cultures for Oakland, Ohewsville , and Monrovia. In the following conside- ration of the plant and climatic values for the four-week growth periods, the four week averages will be used in con- nection with the climatic values for the two halves of each four-week period.. These are given in the two week tables and shown graphically on the two-week graphs. For convenience in reference it may be said here that each culture of the longer growth neriods includes 'he shorter period beginning on the same date and the shorter period next following. T'hat is a four weefejperiod beginning July 1 includes the two two-week periods beginning -July 1 and July 16 respectively.

During the first three four-week growth periods for Oakland the plant and climatic graphs have the relations that would be expected. From the period beginning May 23 to the period beginning June 19 at this station the value of the tem- perature index rises gradually, there is a small decrease in

71)

light intensity and a considerable decrease in evaporation rate, with the plant graph rising sharply, '^he graph of dry

weight then descends from the period beginning June 19 to the period beginning July 3 and ^'^ightly^^rises, from the period beginning July 3 to the period beginning July 16. Neither of these 7;ould be expected from the values of the four-week averages. If , now, the dry weight graph from the period begin- ning May 23 to the period beginniiig July 16 be compared 'vith the two-week evaporation graph from the period beginning Jime 5 to the period beginning July 31 it will be seen that the dry weight exhibits a very consistent inverse relation to the evaporation as shown by the opposite slope of the two graphs from period to eriod. The dry weight at this station sesms thus to be determined during the first five four-week periods largely by the evaporation during the last two weeks of each period. From the period beginning July 16 tc the end of the season the graph of dry weight descends following & corres- ponding downward slope in the graphs of sunshine and temperature indices shown by the four-week averages of these conditions. Evaporation during these -ericds is low and seems not to affect the plants. The sharp downward slope of the dry weight graph from the period beginning Aug. 14 to the period beginning Aug. 2? is probabljr accounted for by the lact that the tem- perature value is very low during the last two weeks of the latter period. ?or this station sunshine and temperature in- dex values show oniy relatively small fluctuations throughout the season and the plants exhibit rather clearly the effect of evaporation.

■'^'or Chewsville, the values of dry weight for the first

so

five periods of the growing season are as in the Oakland cul- tures approximately inversely proportional to the evaporation for th° last two weeks of eaoh of the four-week periods, fhis may be seen by comparing the slope ii recti on of the dry weight graph for the interval from the period beginning May 19 to the period beginning July 14 with the slope airection of the two- week evaporation graph from the period beginning June 2 to the period beginning July 28. The rise in the dry weight graph from the period beginning June 30 to the period beginning July 14 is not as great as would be expected from the I07; ra-^e of evaporation during the last half of the latter period (see the period beginning July 28 on *he two-week graph). It will be noted from the tv:o-week graph, however, that the values of the temperature index and of sunshine intensity are both low for the two-week period beginning July 28 and this probably explains the depresssicn of the dry weight value for the four-week period beginning July 14. From the last-named per- iod to the end of the season the values of dry weight are rel- atively lower than would be expected from the four-week averages. A possible explanation for ea^h of these low values is to be found, however, In the fact that one or more of the three climatic conditions is unfavorable for growth during the last half of the periods. The low value for the four-week period beginning July 6 may be accounted for by low sunshine inten- sity and high evaporation rate during the last two weeks as shown by the period beginning Aug. 11 on the two-week graph. The last two weeks of the four-week eriod beginning Aug. 11 show a very low sunlight intensity value fperiod beginning Aug. 26 on the two-week graph), ^ry weight for the period

81

beginning Sept. 6 is higher than would be expected from the low temperature value due possibly to a relatively high sun- shine intensity during the last tro-weeks of growth. 'R'or the four-week periods beginning Sept. 22 and Oct. ? both stin- light and temperature index values are low for both of the two- week periods constituting eadh four-week period.

?or Monrovia, the dry weight graph rises to a high value for the second period of the growing season in spite of high evaT)oration rates throughout the ^oeriod. From the second per- iod to the third period the graph of dry weight descends, probably following a corresponding decrease in sunshine inten- sity shown by the four-week graph. The still lower v:-.lue of dry weight for the period beginning June 29 m'-.y be explained by the high evaporation ra+"e for the two-week period beginning J;ly 13. The minimum in the dry weight gra^h at the period beginning July 13 would be expected from the low temoerature index value and high evaporation rate for the last two weeks of this period fperiod beginning July 27 on the two-week graph). The rise in the dry weight from the peridd beginning July 13 to the period beginning July 27 is r)robably related to the relatively high temiierature index value during the last two weeks of the period beginning July 27 although evaporation is also high and light a little below the ^easo-^al average fsee the period beginning Aug. 10 on the two-week graph). ?'or the four-week period beginning Aug. 10 dry 7;ei;;;'ht shews a relatively lew value as woiald b-^ expected from the very lov; light value of the two-week period beginning Aug. 24 which is the last two weeks of four-week period in question, fi'or the four-week period beginning Aug. 24 dry weight rises in spite

82

of a large decrease in the temperature index value but the high value of sunshine intensity during the last two weeks of the period may account for this behavior of the plant. The l^t two four-week periods of the season are characterized by low light and temperature index values throughout, which may accotmt for the low plant values shown oj these neriods.

For College, the four-week climatic values do not seem to account for the lov; growth rate for the period beginning June 19 nor for the extremely high growth rntes for the periods beginning July 17 and July 31 respectively. The first-men- tioned r)eriod would seem to have good growing conditions fhigh temperatue index value and low evaporation rate while the last two have about the same temperature as the first with high evaporation rates). Here the climatic conditions during the seconi two weeks of growth do nOt seem to give any suggestion as to why the culture started June 19 should show such a lo-^ rate of growth in comparison with the cultures started July 17 and July 7>1, Several noints worth calling attention to come out of a comparison of the two and four-week graphs for the cultures started July 3, July 17 md July 31. It will be noted that the two-week culture started July 3 is relatively high in leaf -product (which is proportional to dry \-eight) but that after four weeks of growth these same plants shov a relatively low dry weight. An explanation for this is sug- gested by the sharp rise in evaporation rate from the first to the second two weeks of this culture oeriod, /'periods be- ginning July 3 and July 17 on *"he two-week g:^aph1. Also, for the series of two-week culturs, +:he one started -Tuly 17 shears the highest leaf product for the season at this station, but

8;i

it will be observed that after four weeks of growth these plants sho'.^ a lower dry v/eight value than the four-v/eek plants of the culture starte 1 July 31 in spite of the fact that the two- week plants :f the latter culture kave much lo-'er growth values than the tv-'O week plants of the former culture, (see the two- week graphs for the two cultures in question). It is therefore evident that some influence operated during the second two weeks of growth 6f ea^h of these cultures which tended to depress the rate of growth of the plants started on July 17 and accele- rate the rate of growth of those started July 31. The two- week climatic graphs show that the rate of evaporation was higher during the second two weeks of the four-week growth period beginning July 17 than during the first two weeks, and that the evaporation during th© second two weeks of the four- week period beginning July 31 was lower than during the first two weeks. The shifting of the growth maximiun in th© two sets of cultures from the culture started July 17 for the two-week plants to the culture started July 31 for the four week plants thus receives a possible explanation in variations in the evapo- ration rate during the second half of the longer culture periods. While it is true that the changes in ev poration rate upon which the above conclusion is based are slight, it is probable that with high rates such as are found in these three periods, evaporation may be at a critical point for the plants and slight variati-^ns in this condition may very well produce rela- tively large plant effects. The decreasing walues of dry weight at this station for the last four periods of the season seems to be most reasonably interpreted as related to the cor- responding decrease in temperature index values.

84

For Baltimore the two-week cultures all show relatively high values for th-^ first eight periods of the growing sea- son. The seedlings of these periods were thus well along in their growth at the end of t^^o weeks, and the climatic condi- tions during the first half of the culture periods seem to have had a relatively greater influence on the -^lants here than at stations where the plants were small at the end of the two- week periods. The four-week plants at this station, and the four-^veek climatic averages thus correlate satisfactorily, as will be evident from the following examination of the gra^^hs. The graph of dry weight slopes upward with the temperature index graph from the period beginning May 14 to the period be- ginning June 26 and at the same time the graph of evaporation is descending. For the petiod beginning June 25 good growing Gonlttions ( high temperature index and low evaporation rate) would seem to account for tide relatively high olant value. While the sunlight conditions become less favorable from the period beginning May 14 to the period beginning Jujie 25 the influence of the other two climatic conditions would seem to outweigh this. Prom the period beginning June 25 to the period |[eginning July 23 temperature values and sunlight remain nearly constant while the evaparration rate tises. The dry weight graph falls as would be expected. For the period beginning Aug. 6 the temperature index and sunlight are at about the same values as for the preceeding period, but evaporation is lower and the plant graph ascends to a high value, i^rom the period beginning Aug. 6 to the end of the season, the low temperature and light valr.es probably account for the rapid decrease in dry weight production shown in the graph representing the value of this

85

quantity.

The Darlington and Coleman four-week climatic avera.^-es and the plant graphs correlate well, probably for the reason given in -^he ciBe of Baltimore. For Darlington, temperature index value remains praotioally constant for the first three periods. The graph of dry weight descends from the period beginning May 15 to the period beginning May 50 following a corresponding decrease in sunshin*? intensity but the slope of the plant graph is not steep probably since evaporation rate is decreased at the same time. From the period beginning May 50 to the period beginning June 13 dry weight production rises a little although sunshine intensity falls again. We may suppose this to be relate! to the fact that the rate of evaporation also decreases thus counteracting the effect of the decreased light intensity. The period beginning June 26 shows very good conditions and the plant graph rises again. The period beginning July 10 also shows very good conditions, high temperature index, high light intensity and low evapora- tion rate and the plant graph rises to an extremely/ high value. It seems impossible to relate the four-week climatic averages for the periods beginning July 10 a .d July 84, or the condi- tions during the second two weeks of these oeriods to the de- crease in the plant growth rate from 'h- first oeriod to the second. Eor the last three periods of the season dry weight decreases wit|i the decreasing temperature index value and sun- shine intensity as would be expected. The graphs for Coleman are quite consistent with the assumptions made in regard to the effect of the three climatic factors. The rate of dry weight production rises as the temper'^tuire index values Increase from

the period beginning May 13 to the period beginning May 28, evaporation rate and sunshine intensity remaining nearly constant; rises again, but only slightly, from the period beginning May 28 to the period beginning June 11 as would be expected from the fact that sunshine decreases considerably, the rate of evaporation decreases and the temperature index rises dur- ing this interval. Prom the period begimning June llto the period beginning June 24 the dry weight graph descends although the temperature index value and light value both rise. This is ptobably due to the high evaporation value, The evapora- tion graph rises in this interval only slightly, however. The temperature index value remains nearly constant from the per- iod beginning June 24to the period beginning Aug. 5 while the plant values behave as follows: from the period beginning June 24 to the period beginning July 8 the plant graph de- scends, while the evaporation graph rises; the rise in the plant graph from the period beginning July 6 to the period beginning July 22 may be related to the decrease in evapo - ration rate; and from the period beginning July 22 to the period beginning Aug, 5 the considerable decrease in the value of dry weight does not seem to be accounted for b^- the climatic av- erages. There is, it is true, a fall in light intensity, during this interval but it is hardly great enough to explain th be- havior of the plant graph. That this culture may have been abnormal is indicated by the fact that the dry weight gr .ph rises from the period beginniAg Aug. 5 to the period beginning Aug, 19 although t e temperature index value for the latter period is very much lower than the temperautre index value fot the former. The period beginning Sept. 2 shows a high relative evaporation rate and a low temperature index value.

87

Low temperatures and light Intensities probably aeooxmt for the low values of dry weight fol* the last two periods of the sea- son. The temperature record for this station stops at the period beginning Sept. 2 and the light record at the period beginning Aug. 5, however. For Ooleman, as for College, a comparison l:etween the plants for the four-we^k and the tv.'0- week periods brings out several interesting relations between evaporatioh during the last two weeks of growth and the be- havior of the plants. An inspection of the two week graph for Coleman will show that the values of leaf-produot for the periods beginning June 24, July 8 and July S2 are very high, while after four weeks of growth the graph of dry weight in- stead of sho?7ing high points for these neriods is would be ex- pected from the two-week values, shows relatively lew points. Also, the culture started June 11 -hich shows a relatively low value for the txo-week growth period has the highest dry weight of the season for this station after four weeks of growth. This behavior of the plants may be explained by the evapora- tion Values during the last two weeks of the four-week growth periods mentioned. It will be noted that for the four-week culture beginning Jxine 11 the evaporation during thejlast two . weeks is relatively low while for the cultures beginning June 24 and July 8 the opposite condition obtains. The plants of the first-named culture were thus exposed to a lower aver- age evaporation rate during the last two weeks of their growhh than those of the two last-named cultures which probably accounts for the high dry weight of the first as compared to the last two. The culture started on July 22 shows a dry weight above those beginning June 24 and July 8 and it will be noted from

88

the two week graphs that the plants of the earlier culture were exposed to a lower rate of evar)oration during the last two weeks of their growth than those of the two cultures start- ed later in the season. It would seem that here, as in the case of the cultxires at College, we are dealing with effects of evaporation which arn relatively ~;reat since this condition is at a critical value for the plants.

For Easton, the dry weight remains constant from the period beginning May 8 to the period beginning May E5 although there is a considerable rise in temperature. This may be related to the rise in the rate of evaporation during this in- terval. A decrease in evaporation rate would seem to account for the rise in dry weight from the period beginning May 25 to the period beginning June 8. No correlation can be found be- tween the low value shown by the plant graph for the period beginning June 22 and the climatic conditions either as shown by the four-week averages or for thalast t^fjo weeks of this culture period since these were such as would be expected to produce good growth. The culture beginning July 6 is low, rela- tively, but this may be due to high evaporation during the last two weeks of the culture period (see the period beginning July 20 on the two-week graph). The dry weights of the cultures beginning July 20 and Aug. 3 are also relatively lov;, for th" same reason perhaps that has been given as accounting for the low weight of the culture beginning July 6. Prom the period beginning Aug. 17 to the end of the season, the low values of dry weight shown by the plant graph may be considered as due to low values of temperature and sunlight.

811

The four-week evaporation graph for Princess Anne is incomplete, the data for the periods tr^glnning May 26, June 8 and June 23 not being available, and this renders it diffi- cult to explain the behavior of the plants during the first part of the growing season. The relative temperature index values for this station are high and the eviporation values available for the station all low except the one for the first period of the season. This combination of low evaporation rate and high temperature value would be expected to produce higher rates of growth than were actually shown. The unusu- ally low values of simshine may be related to the fact that the plants failed to show higher dry weights at Princess Anne. It will be seen from the four-week graph that sunshine inten- sity is well below the seasonal average from the period begin- ning June 8 to the end of the season. The behavior of the dry weight fot the first four cultures cannot be satisfactorily explained in the absence of the evaporation data for three of the culture periods but from the peiods beginning July V to the period beginning Sept. 29 inclusive the dry weight and the temperature index values seem to correlate as ^TOuld be expected since evaporation is too lov; to affect the plants and sunshine remains at a nearly constant value during this time.

All of the plants discussed up to this point were grown In the exposed statins, that is, in the open with no covering other than a screen of wire netting(of large mesh to protect the plants from injury, At three of the stations, Oakland Baltimore, and Easton.as was noted in "he introduction to this paper, a series of cultures was also grown under glared cold-frame sash supported three feet above the ground, these

cultures being designated as the Oakland, Baltimote, and Easton covered stations. The behavior of the plants grown under glass is very different from the behavio-f' of those grown in the open and a description of grwwth as it took place in these covered stations will now be given.

The cultures grown under glass were placed near the ex- posed cultures at ea^^h of the three places mentioned, so that the climatic conditions for the two would be practically the same except for the effects produced by the glass. That these effects werfi considerable is shown by the very T.arked differ- ences between the plants under the glass and exposed plants growing only a few feet distant. The addition of the covering had in fact as great an effect as would have been produced if the plants had been grown in another part of the st^te.

Only one of the three climatic conditions dealt with, evaporation, was measured for these covered cultures, and we thus have no exact notion as to what sort of climate existed under the glass. The most that can be done is tD compare the plants of these cultures with the exposed plants and call attention to such peculiarities of the covered stations as seem to be general for all of them. We may be sure, however, that the climatic conditions under the glass differed from the climatic conditions for the exposed plants in certain defi- nite ways. Meaaurements show that the rate of evaporation for the covered stations was considerably greater than for the ex- posed as will be seen by comparing the values given in the tables. We may be certain, also, that some of the incident light was ab- sorbed by the glass and that the light intensity under the glass was thus less than the intensity of the light falling on the

exposed plants. Also, we may be reasonably sure that the tem- perature under the glass w-s somewhat higher than the tempe- rature outside, especially on quiet days when circulation of air would be slight with a consequent slight tendency toward equalization of temperatures under the glass and temperature outside. In considering the behavior of the covered cultures, it will be borne in mind, then, that the evaporation is known to be higher and the light intensity lower for these than for the exposed stations while the temperatur'^^ is probably higher for the covered than for the exposed.

The effect of the glass was shown by the plants in two ways: (1) growth was always greater for the covered stations than for the exposed, and (2) the plants of the covered sta- tions showed a noarked difference in m.-inner of grorxth from the plants of the exposed stations. The greater growth of the cov- ered plants was shown In some cases by one, in some cases by two or by all three of the growth measureraents taken. Not only do the plants show greater growth, but the maxima in the graphs of the various growth measureraents for the covered plants do not usually coine at the same times as the maxima in the cor~ responding graphs for the plants grown in the open. The prin- cipal effect of the covering on the way in which the plants grow is shown by a disturbance of the relation between dry weight and leaf area. In previous discussion of this relation for the exposed plants it was noted that the relative dry weight and leaf area numbers are approximately the same for the four- week plants. In the case of the covered stations, on the other hand, every culture shows relative leaf area higher, usually very much higher, than relative dry veight. The stem height in the

covered cultures usually shows high values as compared to the corresponding exposed cultures. The tendency noted in previous discussion for this growth rate to fall off relatively, as the plants b-^Gome larger seems to be to a great extent not active here. The consideration of the covered cultures in detail will bring out the features mentioned above. It should be noted that the culture periods for the covered and expased plants correspond to within a day or two. That is, the exposed cul- ture^and covered cultures were started at approximately the same time and thus extend over practically the s?me growth periods. In some cases exigencies of the experiment made it necessary to take measurements on the pl'^nts of the covered cultures on the day prec^eding or the day following the one on which the exposed cultures were cneasured. This, however, would not introduce ancugh of a difference in the measurements to interfere with the general comparisons here made. In the com- parison between the growth for the exposed and. covered cultures, no attempt will b "^ made to account in detail for the differences between the two sets of plants in terms of climatic conditions, since the climatic influences acting on the covered plants are not accutately kno-^i. After the peculiarities of the cov- ered plants have been pointed out, however, an explanation of their general behavior will be given which seems most probable in view of all the facts of the investigation.

The covered and exposed cultures for Oakland iiffer le3S than the two corresponiing sets at the other stations fBaltimore and Easton) , but show, nevertheless, the general features out- lined above. The plants of the two-week covered cultures for Oakl-?.nd exhibit a much higher value of the leaf-product than

9;^

the corresponding exposed cultures for the perbds beginning June 18, July 2, and July 15 and the stem height is greater for the covered station than for the exposed station for the periods beginning June 4, July E and July 15. The highest value of leaf-product occurs in the period beginning July 15 for the covered and in the period beginning July 16 for the exposed two- week plants. Both sets of cultures show two maxima, in the plant graphs but these are much higher in the case of the covered plants than in the case of the exposed. In the four- week graphs of the covered plants, leaf area is higher than dry weight for the whole season, while in the expos- ed plants, the granh of leaf ar'?a is well below the grai^h of dry weight from the period beginning May 23 to the period be- ginning July 16 inclusive. The maximum for all the growth measurements of the four-week exposed plants occurs in the pertod beginning June 19 while in thej-^overel set of cultures the maximum occurs in the period beginning July 2. Also, the graphs of the four-week plants all exhibit higher values than the graphs of the two-week plants for most of the culture per- iods of the season. This is especially true of the leaf-area. The effect of covering the plants with glass would seem to be to produce a relatively high rate of leaf expansion, and this in spite of the fact that evaporation is somewhat higher for the covered than for the exposed plants.

The two-week plant data for the covered station at Baltimore are plotted to a scale one-half as great as the scale used in plotting the exposed plant values on account of the high values of height and leaf-product shown by the covered culture beginning July 9. The values of both leaf-product and

Stem height for the plants of the Baltimore covered station are above the corresponding values for the exposed station. Also, the tendency of the plants under the glass to elongate relatively more han t e exposed plants is shown by the stem height values for the Baltimore covered station for both two and four weeks of growth. It is interesting to note that the covered plants do not show high values of the plant growth rates for the period beginning Aug. 6 as do the exposed plants. The four-week graphs for the covered plants show very well the tendency of leaf area to reach values relatively higher than dry weight, the leaf area grat)h being well above the dry weight graph for the entire season.

For Easton the covered plants, as compared with the ex- posed, show the general tendencies noted above. The two-week growth rates of the covered plants are higher than those of the exposed plants especially stem height. It will be observed that the maximum growth for the season in both the covered and exposed two-week cultures occurs in the petiod beginning Aug. 3. The four-week olaAts of the covered cultures sho'-: leaf area relatively higher than dry weight. The values for the culture period beginning June 22 are low for the covered cultures as well as in the exposed set. It was noted in the discussion of the latter that the climatic averages and these low values did not correlate. Since the covered plants show low values as well as the exposed. It is likely that we are dealing here with some climatic effect, however, and not with an abnormality of the cultures.

The difference between the behavior of the plants under glass and those in the open should be attributed, it would

primnrily ');

/ 'j>

seem^.to difference in lieht conditions for the two sets of cultures. The 'A'aight of the four-vreek plants under glass is, in most cases, greater than the weight of the plants grown in the open in spite of the higher evaporation rate experienced ty the covered plants. If it is assur:ied that there is a high- er temperature under the glass, the increase in dry weight pro- duced receives a possible explanation. Also, an increase in growth In length of the plants wen placed under glass is ex- actly what would be expected if temperature is high and light intensity low as compared with the values of these factors in the open. The relation between stem height, temperature, and light has heen sufficiently discussed xinder the exposed sta- tions, and will not be repeated here, but it will be seen tha"^ the behavior of the covered plants supports the assumption made as to the relation of this particular growth rate to the en- vironmental conditions here measured. The fact that the leaf area of the covered plants is high in comparison w'th their dry weight may be considered as due to a lowering in amount of dry matter produced b;-. photosynthesis per xuiit of leaf area. Such a lowering might be expected if the light available for photosjmthesis is cut down by interposing between the plant and the light source a screen that absorbs a part of the rays. This explanation of the behaviour of the plants is, however, only an assumption and cannot be -oroved from the data at hand.

Whatever may be the explanation of the behavior of these plants, the facts as they stand show very clearly that the growth of plants under glass is quite iifferent from growth in the open, and the indication here is that the glass acts on the growth rates directly by screen ing out part of the light.

96

and indirectly as well, by affecting the other climatic condi- tions. The habit of growth of the plants is thus altered and probably the amount of photosynthesis per unit leaf area. Such effects, if they are general for plants grown under glass, would obviously be of importance in their bearing on plant physiological experiments conducted in^reenhouses. It would seem that growth behind even a single thickness of ordinary glass is quite different from growth in the open, and that in applying conclusions drawn from greenhouse experiments to plants grown under outdoor conditions this fact woiild have

to b e taken into account.

"'he Forest Station.

The nine series of four-week cultures grown in the open and the three series grown under glass include all the soy- beans of the experiment except one series which was grown in the woods near the Laboratory of Plant Physiology at Baltimore. This series, the Baltimore Forest Station, was located, as has been noted, about 150 yards from the exposed and covered cul- tures at Baltimore. The behavior of the plants grown in the woods seems to support very clearly the assumption upon vrhich the behavior of the exposed and covered cultures is explained. Only the evaporation was measured for this station but the suji- shine intensity was, of course, very low due to the shading and screening effect of the leaves of the trees above the ex- perimental plants. The temperature would also probably be considerably lower than the temperature experienced by the exposed and covered plants. The modification of growth habit in the case of tfie forest plants is very striking as can be seen by an inspection of the r)lant graph for the Baltimore

97

F<jrest Station, The soy-beans, short erect growers in the open, and erect under the glass, but with rel .tively long stems, becone runners in the forest. This effect on stem growth, which obviously cannot be explained by the temperature in the case of these cultures, is relatively very great, the highest alue for the two-week stem elongation being over 4-1/2 times the seasonal averag-, and the highest value for the four- week stem elongation being a little less than 4-12 times the seasonal average. The four-week plants also show the same - change in the relative position of the leaf area and dry weight graphs that was shown by the covered plants. The leaf area graph is above the dry weight graph for the entire season. The forest cultures thus show a similarity to the covered cultvires rather than to the exposed. This may be most reasonably ac- counted for by assuming the similarity in the behavior of the plants to be due to a similarity in the light conditions for these two sets of cultiires.

In the introduction to the discussion of the plant data, it was stated th-^t the main purpose of this experiment was to test the use of plants as instruments for me-^surlng climate. 5he feasibility of this method of measuring climatic condi- tions has been urged by Livingston and McLean and by McLeam. "

V McLean, F. T. , A preliminary study of climatic conditions in Maryland, as related to plant growth. Physiol, Res, 2; 129-208, 191V,

These authors state that the plant growth rates are a result of the total of all the environmental conditions actins? on the

9H

plants and that growth rates are therefore, -from this point of view, themselves a measure of the ability of the environment to r)roduce growth. It is assumed that such a method of measur- ing the environment is susceptible of "standardization". That is, certain plants may be selected for use as integrating in- struments, their growth pe^uliarites studied, and conditions defined under which their gro'A'th rates would measure this or that environmental condition or complex of conditions. It is suggested that the measurement of environment in terms of the growth of standard plants, assuming that it is possible to standardize them, would in all orobability give results applicable directly to many purely scientific and practical problems. In the present exr)eriment an attempt is made to grow soy beans \mder experimental conditions so controlled that the growth of the plants constitutes a measure of a cer- tain part of the environmental complex ordinarily termed cli- matic, the environmental factors here measured being tempe- rature, sunshine and evaporation.

From this point of view, every gwowth rate for each of the cultures, both two- and four-week is a measure of the climatic complex made up of the temperature, sunshine and evap- oration conditions acting on the pl=ints during the period for which the cultures under consideration was growing. It is, of course, realized tha "^ in the present experiment the con- trol of the environmental factors other than the ones to be measured in terms of the growth of the plants was far :^rom complete, and tha"- the above statement Is thus only approxi- mately true. Assuming, however, that the control of conditions other than the one to be measured was such that the growth

of the plants was mainly determined by temperature, light, and evaporation, it may be said that the peculiarities of the plant graphs for each of the various stations are measure- ments of the seasonal climatic complex for that station as this is registered by the growth of two-and four-wek old soy- bean seedlings. The (ggj'O:**^ graphs represent graphically the "readings" of the standard plant when groTvn in the ?-'ay de- scribed and exposed to this seasonal march of conditions. The reading for each r)eriod m^ty be regarded as the "plant produc- ing power" of the climatic complex if the growing period. It has been previously brought out that the readinrs on the stand- ard plants seems to bear some relation to the values of the three climatic conditions dealt with, especially in the case of the four- week plants.

If we average the readings for the season at each of ^he stations, we get a number which represents the average seasonal value of the plant producing power of the climatic complex as it is registered by each of the growth processes considered. ThJ s has been done for the exposed stations for the two-week and for the four-week periods with the results shovm in Plate ZIII. The ordinates of the graphs of averages represent the average daily relative growth rates for the season for the two sets of plants and the average daily relative intensities of the climatic conditions . The name of the station at which each average growth value was registered is given immediately be- low the ordinate on which the value i.s clotted. The total plant producing power of any station for either the two-week or four-week grov/th periods would be secured by multiplying the averare value of this growth rate during the season by

the length of time (expressed as the number of tv;o-week periods^ during which it was acting. The length of time during which the average growth rate is active would be the length ':^f the growing season, which, for soy-bean and most agricultural plants, would be the length of the frostless season.

The plant producing power of any station may thus be re- garded as equal to the product of two factors, CLn intensity factor and a duration factor, the first derived from the read- ings of the standard olant and the|second the length of the frostless season. A large value for either of both of these factors at a given station would result in a high plant pro- ducing power and a low value for either or both of them would re- sult in a low plant producing power.

The two-week seasonal averages for the nine exposed sta- tions of the two growth measurements taken represented graphi- cally in Plate XIII show that the plant producing power of the climatic complex is about the same whether it is measured by stem height ot leaf-product. The range of variation for the former measure/pent is from 76 for Monrovia to 125 for ■Baltimore, In terms of stem height thus the average intensity of the Monrovia climatic complex is 61 per cent of that of the Baltimo e complex. Similarly, leaf-product varies from a minimum of 77 at Oakland to a maximtim of 119 for Baltimore. As measured by leaf-prodiict therefore, the Oal<land climate is 65 per cent as efficient as the Baltimore climate. If we in- troduce the duration factor, however, the plant producing power of Oakland will be still further reduced since the grow- ing season at this station ts short, including only 9 two-v/eek periods as compared to the lengths of the growing season at

ioi

the other stations which include 11 or 12 tv,-o-v/eek periods.

When the two-7.-eek averages are compared with the four- week averages, it is seen that the graph of leaf area and dry weight for fout^^weeks do no+- parallel the grar)h of leaf-pro-

duc t for two weeks. That is, the climate, registered in one

tv/o-week way by ^leaf-product fwhich is approximately proportional to

leaf area and dry weight) is registered by the four-week leaf area and dry vreight in quite another ay. It should be remem- bered that the two-week and the four-week cultures were carried on over practically the same interval cf time at each station so th.at the plants of both series are registering the same total climatic effect. It will be noted tha"^ the two week stem height averages parallel the four-week stem height aver- ages rather closely, however. This behavior of the various growth measurements as it bears on the use of plants for m,easuring environmental conditions will be discussed below.

As has been noted above, if plants are to be used for the measuremant of environmental conditions, it would seem neces- sary to standardize the plants. It is known that a plant is always changing in its power to rescind to external conditions. Also, +^here is a change in *-h.e internal condition of the plant resulting in a decrease in the ra-^e of growth as the nlant be- comes older. Both of these points have been emphasized in previous discussion. The relatively great effect of evapo- ration in the second two weeks of growth on the dry weight pro- duced is an illustration of th"^ first type of variation whJle the gradual falling off of the rate of growth of stems as the plants become older is an illustration o+ the second type. By starting all the plants from the seed, as vas done in the

102

present experiment, we may be reasonably v-ure that the zero point of all the integrating instruments 's the SToe. At the end of the growing period however, the "setting" of th^ instru- ment is not known, since we do not know what cycle of changes the internal conditions of the plant have gone through. In order to apply the data secured from standard plants to plant growth in general, it woiJd seem necessary to correct the recdings of the s ' andard plant in the same way that the readings of any physical instrument are corrected for instrumental varia- tions. At present no way is known by which a number can be given to the reacting ability of a plant or to the stages of the cycle through which the plant passes as it matures. Variations in the internal conditions of the plant such as these here noted must be responsible for the fact that the seasonal avera es of the two-week leaf-product value, a plant measurement propor- tional to the leaf area and dry weight of the two-week plants, registers a given set of climatic conditions different from the way in which this set is registered by actual leaf area and dry weight of 'he same r^lants grown for four-weeks.

Several methods suggest themselves by means of which the difficultji noted above may be surmounted. One possibility is to evaluate the internal conditions of the plant and correct the readings for changes in those conditions. This method would involve a detailed study of the standard plant and at best would seem to promise a solution of the problem involving com- plicated mathematical reiuctions of the plant readings. There is every reason to believe th .t a correction formula for plants considered as integrating instruments would, if obtainable, be exceedingly complex. Another, and apparently more honeful

method woxild be to find some plant measurement or to evaluate some plant process dependent on external conditions and at the same time not subject to extreme variations lue to inter- nal changes in the plants. Or, a measurement or Drocess might be found whose variation due to internal changes woiild follow some simple, easily determined relation. The fact was noted above that the graphs representing the average seasonal stem height values for two and four weeks of growth were parallel in general. This measurement may, therefore, have the quali- fications noted above. That an approximately constant rela- tion exists between ^s^ t-o and four-week stem height may be shown by dividing the two-week seasonal averages for each sta- tion by the four-week seasonal average. We find that tMs gives values of the ratio as shown below;

Oakland - l.Pl

Ghewsville 1.16

Monrovia 1.18

College 1.19

Baltimore 1.20

Darlington 1.07

Coleman 1.20

Easton 1.27

Princess Anne 1.15

The average of these ratios is 1.18. If, therefore, the stem height of the soy-bean, grown as a standard plant, be used as a measure of the climatic complex and the measurement be express- ed as relative aver ge daily increments as in this study, the two-week readings may be corrected to the four-week readings (considering these as the standard) by dividing each two-week reading by the constant 1.18. Evidnece which cannot be given her'= shows that the rate of photosynthesis also, is dependent

on the three climatic factors here dealt with in a definite way and apparently not subject to large variation due to in- ternal conditions of the plants. Another possible method by means of which to elimin ite the internal peculiarities of the standard plant would be to grow the standard plant and plants whose reaction to external conditions was to be expressed in terms of the standard ujider the same set of environmental con- ditions and to compare the growth rates. In this way ratios between the standard plant and the plant 'vhose reaction to the given set of environmental conditions is unknown would be se- cured. We might then grow the standard plants in various en- vironments and predict what growth would be made in those en- vironments by plants whose relations to the standard plant were known.

105

Conclusions .

The main resv-lts of this study may be summarizec under the following hesds; (1), the interrelptions of the climatic measure-, ments, (2j the interrelations of the plent measurements, (5j the relations betwec;n the plant and climatic values, and (4) the measuremient of the climatic complexes by soy-bean, considered as a standard plant.

The principal features of interest in the clim.atic data are the correspondence of the sunshine ena evaporation graphs ana the slight variability of the physiological temperature index as comiparec. to the indices of sunshine and evaporation. ?rom the first of these it mry be concluded that the energy absorbed from the sun's rays by the porous cups plays a large part in the determination of the evaporation rate. ?ror;: the second, it would seem that during any £;iven time interval the air temperature at the various stations here dealt with was less subject to local variation then either the sunshine intensity or the evaporation rate .

The relations between the plant measurements fall into two divisions. (1) Ihe relation between the elongating tendency of the plants and their tendency to expand their leaves, and (2> the relation between the leaf area and the dry weight. The data here given indicate that the rate of elongation is greater than the rate of leaf expansion when both are small, ana less than the leaf expansion rate when both are large. Dry weight and leaf area are so related in soy-bean that the relative values of these measure- ments are approximately numerically equal. Both of the above con- clusions apply to the plants of the exposed stations. For the coverea stations, the value of stem height is relatively large

106

eno- the relative value of dry v.eight is always less than the relative vf lue of leaf area. The two latter stateTr.ents also apply tiQ the plants of the Baltimore forest station.

From the data presenteu in the preceeding pages, the follow- ing conclusions may be drawn as to the relations between the climatic conditions and the various growth processes* (IJ The height of the plents, in so far as it is determined by the con- ditions here dealt v.ith, seems to be influenced m^ainly by sun- shine ana temperature, in the CFse of the exposed plents the tem- perature seems to have r- preponderating effect with sunshine acting secondarily. The effect of temperature is direct, high tempera- tures accelerating growth in height and low temperatures retarding it, while sunshine has an inverse effect, high values retarding and low values accelerating the ster. elongation rate. In the case of the plants under glass stem, elongation rate is relatively high ana in the case of the plants of the forest station it is relatively very high. The facts indicate that low sunshine in- tensity produces some etiolation in all the plants and consi- derable etiolation in those grown under conditions such thrt a part of the incident light is absorbed before it reaches the plants as is the case for the glass-covered plants enc. those grown in the forest. (2j The relation apparently existing bet- ween the dry weight of the plants (or the leaf-product which is an index of dry v.eight in these plents; and the climatic condi- tions here dealt with may be stated in the form

r = ^LJliKih)

i:i which the symbols have thr; meaning previous].y given.

ov

(c; The fciir-week plants seem, in r.eny cases, to be especially- susceptible to high evaporation rates during the last tv.'o weeks of their growth.

Considering the soy-bean as a standard plant for the measure- ment of the efficiency of the clim.ate in various parts of the state of l-.^arylend to produce plant growth, it would appear that during the season of 1914 the three stations in the western part of the state (Oakland, Chewsville, and Monrovia) showed a lower plant producing power than the remaining six stations, the term plant producing power meaning here the effectiveness of the climatic complex to produce growth in young soy-bean plants. It would also appear that the climiPtic comiplexes of certain culture periods weremore efficient for growing soy-bean than the averages of the environmental conditions would lead us to expect, while for certain other periods the clim&tic com,plexes were less effi- cient tc produce growth than would be expectec fror the averages for the periods. Taking all the facts of the investip-p.tion into accoui&t, this would seem, to be due to the v. ay in which high and lov. intensities of the conditions were cistributed throughout these periods.

108

^- ^ceK periods.

JUMF JUMC 19

.lUriE JULY JLH-^ JS It. ' 'i

3'

•\uCi .>U0 SEPf -1,1,. c.EF-T 5EPt

Av.

Culture number

1

2

-5

•1

S

0

/

0

9

Lenqfl? opqrowincj period , dai.^&.

li

1-4

H

13

15

1-1

1 J

1 0

»i

iHurnber of ^slarcts.

^

.S

5

t>

i

0

to

t>

*;

fTernaindef summation iricie)^.

-'O-l

i54

■«"

3e.^

-131

3oS

5-13

3TO

.2B£

3b 1

"temperature inde^

.0

Dt>

a5

TO

97

fc.7

6-1

(3fo

4i

S9

Avcaqc duil^ mean t^mpcraturc.cieq F

65

to-l

fc-a

bT

fci©

(=3

tjfc.

faZ

feji

65

Ai^craqc da.lu relo+i ve cvaporat-ion

15 J

13^

96

79

104

90

71

J7

fe9

ct6

Awcr'oqv! duiiu relative suushiiJc

tZ-£

IZ2

109

103

t ifa

tos.

1 10

a-5

o«

105

AverocjC rfuiU relatii/e. increment" in

^&

9&

lE-l

S-^

9fc.

96

...

5i

-42.

ao

Avcroqc dailu relative increment" in lca| - pi-octiic:^

-

ai

e?

7-q

39

70

ez

-le

-

7 7

C)-\L VV5VILLL e v^veK periods ExPO^eD STATION

^7777 JUMf

JUME 10

junf JUNF 30

junr 1-J

JUL1

~0

AUG. £5

AUG. ^s .tPi a

5f.Fl i2

OCT 1

Oill OCT.

AV

Culture number

)

E

3

•4

3

<o

7

s

q

10

1 1

Lcni^th o[ ^rowiuq period, Uu^jS

1-4

,.

,->

,.

i-i

14

|4

i-^

1 H

■'

■'

Number of plartt:^

^

u

-1

"

i

"i

3

5

&

t>

t»

Kiiiouii icte' suiriinahori inde*..

•lit.

^^3

■VIZ.

4J3

52.i

■450

31 7

43to

302

3ae

JOl

4z.a

AvcrciLit; (Jci.l-j rt^lotivc; plJ \j510loq Iccil ten ];iera*'urc tnde-..

9?

.00

1 >9

1 le

'49

<<£

i-qs

■ oi

•^1©

5 '

f-Ji

98

Averoq^. Jad*! mcun temperature, deq Ff

Gq

T 1

1 i

Ti

7o

Ti

7C

TO

01

(51

Gi

©9

Average dail\j relative evuporo+ioi-i i n d e •• .

no

' 15

|07

7S

1O6

eT

105

SO

TQ

73

G>4

91

Avev"cji^c UqiUj relative suashine 1 nti3i-\ipir\j.

l£i

iS*^)

no

■ 0&

i£i

94

qe

fciT

104

73

^■a

98

Average daily relative increiTrent' n? stem Ueicjhr.

9i

93

113

liZ

,.0

'»e

■>a

TCj

je

Z.I

37

87

AvGroi^c OotiM rclotivc- iiicr.zmcnt' 'm leaV- produc-T".

10 -1

..0

17.'

qO

liT

12 7

100

b-2

52

cTO

-i

a4

EXPOSED .3T>\T(Or\	1	JUME JUME IS	jorte 13 juriE	June 2'-) JU1.1 13	13 JUi-t 2.7	if? AU<3. 10	AUtt Ai>o. 2-*	3£PT 7	5EP7	OCT. a	OCT. 19		-^v.
Cu/furc rrumbcr.	1	£.	3	4	3	b	7	e	9	10	1 1
Lfcoqt/i o/' t^roLuittc^ jx;ric»d-,daLjS.	M	11	14	14	14	PH	14	14	IH	) 7	M
liumbcr ot' iJluiitij,	3	-5	b	4	-	5	6	5	fc.	t>	fe
'^cniaiiider suimnation index.	4HJ.	434	47a	•ii^-l	J5&	4GJ>	51 1	43j	3&^	3eo	■^•o^.		451
Aueroqe dailvj relative pbi^sioloq ical tempero+u'e indc> .	lOS	no	12.4	IIZ	ise	117	144	10 J	59	J3	.59	1	104
Averaqc dail^ mean "t'emperafu re .dcq F	71	7 i	73	7 1	77	7£	7J	7D	e>2	&£	03		-TO
\wGra<\^ daily relative cvaporulioo Mjde*.	-	I4&	iS£	9^	M7	leo	ria	OS	90	a^	76		106
'^vsraae dailv rcluTive sun^b.ne	152.	1^2.	I04	loa	MO	lOi	9S	62	'O5	06	ja		100
^vcra-^C Jailv( relative incrc'Ticnt 10 stcTi hctqlit:	73	S'	ȣ'	96	US-	90	10/	76	20	67	25		T9
Averaqc dady relative lnc^en^ent m leaf- produc-r	66	<*o	HO	63	OS	109	104	...	6	J6	4		80

PLATE I.

10!J

COLL.i_Gn a -week pcriQclb

MA1 duME

June jufiE JUME JUL!

JULT 17

17

JUi-1 31 AUGi.

H ^1

.^7 3tFr

to -5 SEPT OCT,

oc r: lO ocr

AV.

Culture nuntbt^r

z

3

■^

-S

^

7

e

9

ro

..

Id

Le.n^t'h of cjroKvina period, days.

n

15

M

M

■■q

IM

u

14

IS-

13

14

Mumbcr of plciv(ti>,

G

S

^

fa

<;

■1

5-

«>

-

6j

to

Rcmuiaclcr sunjmation Index.

5-i<,

40^

i^e

'J'»4

620

H'1'1

47©

^11

:s70

-5-i5

^Z.1

150

Average aa.lN( rcla+ive phv^sioloqical tempcrJi"Mre index.

loa

103

I5T

"31

1^6

,»

143

MS-

G7

JO

S2

lOd

Avcraqc Jail^( m<:xin tempv2rafurG,deq. K

71

70

,.

/I

76

^■^

76

7i

O'l

G(

oZ

TO

Aycraqe dail\j relQ+Wc cvcipora+ion 1 lldc>^-

l»30

IS-i

.23

^C>

143

1-17

las

I07

73

34

©a

11^

-

-

-

-

-

-

-

-

Avcrac^c Ja>l^J rciotwc incrcmen't" m st-cr-n hciqhr

03

|3£

115

lie

l^^

liO

90

7o

4©

33

9^

Avcraqc cloil>j relative incrcmen+in lea \ - proctuc-t:

■9t)

I'd3.

S"

152.

.o.

150

UT

t>£

-

13

lO

1 D\

at)

MAI JUNE lO

ouNE lO JUHE

JUL1 4

JLILI JULI 25

2i ALKi to

AUG AUij. £0

SCP7 3

5Ei=T 5EPT 3 <4 sepr'ocT. 1*-* 1

Ot T, -4

Av.

Cu|-turc number

I

Z.

3

4

J

e.

7

e

^i

lO

"

Lenqth o|-qro^M7q perrod. dui^i,

13

12

15

14

14

14

lA

14

lO

l£

13

riuinbcr oV plunt-b.

^

-

.5

G

&

4

•5

5

"

S

-S

Rcmaiiictcr sum(-naVron index..

461

3 70

?oa

^6'^

340

50t.

5-1 -•■

bOd

5^3

-■■OO

^r ^

i4e

Averaqe doilij rclutiwc pWi|i> i oloq icol tCmpcroture I'lUdl"

91

•^'^

12,3

IE.5

■iSa

,H.

leo

13-J

C2.

G7 |.3

tl£.

AxiCroqe doiUj i-nea^i "tcmpcrutxiix., dcq P

Ci9

70

73

7J

73

7i-

70

rj

OA

6^

o<4

1Z

Awcrac^c dati-j relotSvc evo poraftoi-j

127

lOZ

Its

CO

4a

liZ.

no

ai

t3<>

sa

6H

93

Avorciqc Oo'il-j relative sunshine, in h-njifs^.

137

a^

lO©

60

^^

72.

79

V4

dZ

ag

J'V

—

36 125

Avcaqc Ju.l,^ rclalive incrcmon-t- in

9i

-

14S

ISZ

183

IS^

I5i

1-19

73

50.

90

Avercjqe da,tv| rclotivii increment in

lOo

-

12.5

ie>3

l=»4

1 lO

ZJ3

ISO

54

2.2

4=1

iiq

DARLir^GTOrs rxP03£p STAtlOM

15 30

MAi 30 JUNE 13

JUHE 13 JUNE ze

JUNE JULY 10

10 JULT 2-1

AUG. 7

7 AUG Z|

AUG 21 SEPT 4

5EP1 4 3f^FT 10

SFfT 13 ocr 2

>JCT OCT. le

Av.

Cul+ure number.

I

E.

-

-*

-5

t)

r

3

9

10

"

Leaqth o\ qro«-vi.7q pcr'iod, da ^|i..

1.5

14

i3

i4

14

11

14

14

1-1

14

!<)

^^unabe^ of \jta\Ltz>.

fe

S

0

4

5

5 1 -

.5

-

.5

4

Rcnioiiidcr sui-iii UQtioii ii7dcx.

^30

4iZ

342.

"112

b'Z

4S6

506

412

30Q

3.b 1

2.56

-lii^

Averaqe da^Uj relo+ivG ph\^sioloqical tei-ri pi^rature index-

ar

S9

94

93

149

113

146

123

4i3

59

54

41

Avcroqc doiUi nicon t-emperoTure, clcq. f.

oe

70

o9

TO

7&

72

76

73

S(

<32

w3

6S

Averaqe da. Kj relative evaporation index..

lio

.2J

90

o£

7-1

7T

7q

0.4

32

3C,

3^J

e 7

A^veraqe da. Ivj re\ativc ^i-mshin^ iiTterisiVvj.

154

|Z4

92

T7

1 II

,u

1 1&

IS

9&

9C

43

100

Averjqc da.Kj. r-zlati'C increnicnl" 111 5t Gni hCiqhT,

liO

lOJ

I4i3

111

2a3

t-IG

-

I07

■iZ.

43

SH

.13

Av3.ruq5 da.Uj roloti ve. nicrei nent- 117 laaf- proJuot.

135 ll

IS,<^

15a

11.5

aqi

I-IS

loa

as

-*B

"

lie

PLATE 1 1 .

lii!

COLEMAM EXPOSED STATIOn-

MAY MAY li za MAt JUNE ZS 1 1

JunL II JUNE ^■4

JULY i JULY a 2^

JULY AUu

AUG,. AUG.bEPT 3EPr 3CF7 OCtH .5 .1 ! ^ IG 30 13 1 Aou [sEprpEirstPiJocT. o._ rl * 1*1 ■ E f n> ] 3o I J £tj 1

Cul ture numlper.

1

2

-5

^

-5

t.

7

e

9

1 1 lO II h£ 1

Lenq-l-h op qrowinq period, davjs-

li

1-^

13

14

l<=3

■ '4

i-q

(•4

14

1-1 13 , |.£.

Numberr o|- plo.iti.

6

e

M

^

o

e.

i^

o

1

-1

M

6

ReiiiQ.ntJcr iiuiumation i.iUek.

-"^

^e&

^i3

-lee

5oO .517

S3e

■szo

3^0

.699

-

—

ITZ.

Avei-aqe doiU relative ptiMsioloqieal temperature index..

SI

115

lie

I30

I70

1

i5a

\^e

&*1

©JT

-

_

IZ3

Ave.raqG dailxj rnean ■te.mpera'Hure,deQ. I;

&6,

IS.

eT

i'=i

T?

7G

77

76

ei

e7

_

_

7a

Averaqe daiKi rela+ive evoporation inde^.

139

152

13T

ISO

1^3

ISJ

lic»

9&

)32

I3S

>JT

33

1^7

Avcraqe dat^ rcio+ive suashine. in+eaS)t\t.

i-qa

l&O

1S5

<Si

(•^1

12^

1 IS

&5

-

-

-

—

I20

AveraqcaoiK) reloTivG incrGi-.7Gii1- tn stem heiqnl-

T<^

ei

lOl

13S- l^q 157

12.1

07

<5Z.

53

&7

ze

96

Avci aqc dailsj relative incremeiTt- in laaf - prodf-ict.

lOS

1 1 1

qT

172 £11

ISH

i3a

lO 1

33

53

jia

A

I07

riA5Tors e- ^icek per .ods, EXR05PD STATiOJH

0

MAY JUWE O

a JUNE

junE JUl-t

^O

JULT 3

AU<i 3 Aua

AUii 17 AUG 31

5EPr

5EPTJ5EPT OCT SEfTOCT ot r 20 II ZL.

AV.

Culture numbGC

'

2.

3

4

JJ

to

7

e

9

JO

J 1

rrr

Lcnqth lj{- qrowmq per.od^doL^s

17

1*3

H

I-)

M

,.

M

H

1-1

11

■ 3

1 5

r^uniber o\ plaut=>.

6

^

5

G -I ^

J

»?

5

«

&

3

Rernanclcr 5Lin7nicitioi? indc.-..

<lJu

153

M5I

'^gz

^3Z|iqj

513

ol9

39a

3S 1 ) 32 (3

I 347' 4H5

AvcraCfC Juilsj rcla+itfe phvisioloqicol tempcrotwre index

o7

ll£

iia

'31

154

133

Ml

I4S

07

70

04

1

Avt=raqc dail\( mean +empcrat"ure,deq.F

e5

71

71

74

77

74

7G

It.

ei

&6

G*l

G2.

1 71

Ayer.jqc doil>^ relo+wG evaporation index.

gs

I30

153

91

S<b

>33

lae

113

Ml

I04

Q&

82.

lOS

AvGruqi^ da.U relative sunsDioe .nton3it>,

tqs

riE

m

i20

IZ..1

lOQ

.1 (

->T

=) (

OJ

^■i

52.

.u

Aucro^o doilsj r^laTivii incrcmeni" in stem heiqhh

1^

oi

112

132

l-KJ

lAC

'i2

lis

8.

4i

o5

5*1

95

Averaq^ doil>, relotive increment in leaf- lJ^odoc^

Tl

<n

• m

9T

ISS

I30

)U3

IE9

<3.4

JS^

-

-

i05

^- cuec ^ period::? E>POS>EC ^TATiur-H

r-iJ^\

JUrt£ O

JUNE a JUME 23

JUflE JULl 7

7 JULY 21

JUI.Y 21 AUu 4

AUG -4 AU<i Id

ALhi 5EPT

aepT 3£pr 13

5::pr IS SCP7

SEP 2 J OCT 12

12. OCT £7

AV.

C*jltij<-e ■lumber

'

a

3

4

S

tj

7

£>

9

lO

"

12

Lcnqth o\ qro^livq penod, day5

IS

13

15

1-^

M

14

(4

-4

|4

14

"3

15

Number oy p!aa+s.

e.

-5

tj

fe

fo

t>

3

<3

to

5

5

<*

Remainder S^jmrnottou M«dc»k.

3S9

■\os

504

493

51-1

463

513

SSC

3e7

3t.i.

315

i^S;432.

Averoqc dailv rclotivf pbx^sioloqicul ■tCfn p6f~Q tu re 1 ndc X.

s-^

'05

lOl

132

,.H

I2J

144

l4o

■34

7*

I 1

43

101

Avero^e do.l^^ .nean femiJtruTure.deq.ff

eo

70

TS

T1

70

7^

70

To

«7

o3-

&3

3,

TO

Averoqt duilvi relative c va ijorat.oi > lOdc. '

117

'3-4

-

-

o3

aj

lOi

73

73

/a

70

3.3

a4

Ave ro^c dai l>( relative, si^nshinc inten^it*j.

■ oi

11©

i03

«^

aa

qt.

7 J

o«

rq

a^

ti7

^ii

S6

-^vGrLiqc du.Kj relative i'>c^cn^en^ in 5tem hcic^lJt:

o^

no

107

1Z.4

197

143

155

lai

79

©5

33

■•■5 lOS

Aner-oqe doil>^ relative increment" ir? lea J-- proj^ot-.

9to

,..,

■ 04

' 1 O

,o-.

I02

■ 77

138

82

34

18

c 90

PLATE 111

Ill

OAKLAMD EXPOSED STATtOM

MAT Jurie^juriejUU jum^juuyLiuly Jua 19 j 3 1 16 1 31

AU(S

JULY 27

AUG M s£pr 12

27 5£PT 25

AV.-

Culture number-.

'

s

3

^

S

t>

7

0

Lenqfh of qro*vn7q penoci, dai^s.

27

a©

Z7

za

29

27

29

29

f^unibcr o^ plants.

4

5

-5

e>

3"

6

&

e

Rciiioiiiclcr surniiiahtiii indcA.

esa

755

7fo5

795

eoo

Tl3

Ti4

652

735

AvC'-aqc dail^ relative phv)3ioloqicol ^cmpc.-c^ru^c indc*..

&£.

-?6

oz

S3

77

6e

65

5S

Tl

Averoqc doili^ mean +£1x1 pe rot urcJeaF

g:>

bt.

t,l

&7

67

to5

64

62

65

*^vG^a^c du'lvi rciotive c i-o poroTioo incJcjk.

i^e

il<*

ST

92

97

81

64

G3

94

ize

lie

lOb

110

109

■ OG

=.7

as

1O6

-Uo'-a^c do.l'.j i-elot,ve mci-Gment- in stem hcjqht:

et.

7d

"Jl

e*?

72

ai

83

47

Tl

rcaf" area

4i

67

91

TS

eo

so

■T2

32

Tl

Avcroc^c dQ-l\j i-claf've tucnEmcnl" in d<-H v-,v:.^hV, !

5.

T7

loa

Si

at.

33

T6

46

T9

CMEW5VILLE: -J-wvecK pe'""io--l£>. EKP03ED STATiOM

MAY juriE ifo

JU«E June JO

,junc 11

JUME 30 Z3

AUk, 1 1

JtJi.1 2e 2i

AUQ 5eFi: a

AULi 25 SEPT £2

SEFl e OCT 7

SEFl OCT 20

ocr 7 MOV 3

AV.

CuH-ure tTumber

1

£.

3

4

5

fc.

7

a

9

to

M

Lenq+W of c^ro»^/inq pGr.od, dQ\js..

iB

ze

£e

ee

^e

e©

2a

^e

2*^

20

27

'^i.i'Tibcr of plants.

€>

„

•1

£)

s^

1

3

^

6

^

&

Rcuanidcr- suimnation index

a ST

-,,3

925

976

973

96T

953

73 1>

650

629

4 90

623

/^</>3rCii:\c. dailv relative pti^^sioloqicol vamperatur-G mdex..

99

tl3

lie

131

131

129

IZ4

7Cb

5^0

52

37

*)e

A^enaqe cIqiIm ixicon tijmperotui'eiiieq ff

^o

7C

72

T-1

7^1

74

-T3

t.e,

6r

e.2

58

69

Avcroqe dO'ls) relative evapor a"tioi7 llJCl^^

(12.

,,.

91

92

9S

96

93

eo

7G

69

""

<^o

Averoqe da.W relotivc suitshiae intsasitv).

iZS

IZ1>

...

1 l-=1

loe

96

S3

6<S

39

59

56

9-5

Ave-'oqe do'i\ nela+ive 'nc ■"Ciirent "in stc.ri Dcqnl

73

103

1£2

q7

91

a4

&4

50

47

4.

31

75

Av-3rjqe dail'y relative I'ncrei-nent in \eo\ area.

73

9"

13E

03

S9

94

99

57

44

4^

12

74

IS

I02.

m^

6T

9.:

QG

eo

02

SI

3S

z;4

TO

MONROVIA fXF03ED iTATlOrA

M4V ■ s lunE 15

<JOHt IS JOLl 13

JULY 27

JULY IS Aua to

27 AU6 24

10 SEPT 7

3EPT £t

SEP I 7 OCT e

sEfr 21 ocr (=1

OCT a 2

AV.

Culture number

1

£

3

■^

s

&

7

©

<?

•0

•■

Lenqtb o-j qrovvroq . period. datj3

^e

^e

ea

ee

e.&

26

PS

3'

^a

25

MLMTtbcr ot plonts.

3

3

b>

4

&

6

5

6

6

<3

Kc.inJ-rtdcr SLin7n7ati*oji inde>^.

5^o

132

^32

090

999

97-i

9-*-)

an

"°

0H6

Hto3

BSi

Averoqe daily r-elative phi^sioloqical tem pcrotu re . odc'".

loa

117

.IS

134

137

• 31

125

a2

ife

se

53

lOZ.

Averoqc doil*^ mGon t"ei-npGrQ+ure..d*i£) F

-

7£

72

T4

7.5-

7^3

73

Gfa

fcZ

63

5e

ee^

Awcraqc doilvi relotive Cvoporatio.-i i.idc^

-

l-=l'1

12.3

lOJ

119

llG

90

32

69

7-,

ei

105

Average dai<i\ rclotive sunshine .iitens.tNj

,^-,

115

103

100

lOB

102

90

94

ee

59

so

96

Avcra Qc da-U reJoTive increment i>"» item heiqht-

63

76

e.

QA

70

91

S4

Si=

4-1

4i

za

66

Avcraqc dollu reloli\C increment i<i IcjI ur-ca

71

I08

©a

99

71

104

75^

TO

39

46

12

Tl

Avcroqc da.lu relat'tve IncremGnt in clr^ wGiqljt.

70

I >4

loe

99

39

ICK5

et

B7

4r

4e

24

T9

PLATE IV

m

COLLEGE

junt

^3 JUME l!^

onE junEJULi JuLi to n 3 IT JULI JULI JULY AOQ. J IT 31 M

31 57

AUO. 1-4 5EFT lO

MX. 1 £7 1 2itt7 .^5

5tPt1 lO OCT 10

5£pr OC7: 2'!

Ci-ilt-ure iiLimbGr.

.

2.

3

-

5

o

7

a

9

lO

II

Lcnq+h of c:^rowinc^ period, do^p.

2tJ

S-T

27

CO

es

ae

27

27

29

30

£9

r^urnbcr o^ plai-rV^.

c

G

^

3

€.

■^

-^

&

C.

-

&

FCcmainJci~ SLimma^ion hide*.

oba

9SO

■Ijo

020

Ol-*

"-

•>7 2

727

bar

7' 3

&to.Z

C)L>2

Avcrtiqc doil^J relo+ivc pWijs'ioloqicul tciipc rat Lire indc>^.

as

no

lei

OT

141

(■41

136

12^

91

• 5 >

32

,,>.

AvcrocjC Ja.l>i mcon fcinv>craKirG,dc^. K

CT

71

73

T.>-

7-J>

T5

15

74

o?5

o3

G2

71

AvcfQuc da-Kj rclufiwc cvu|jorat"ion

loq

157

i3q

no

120

145

tA 1

121

93

ai

63

124

Avcruc|c cIqiInj rdutivc sunshine intcn&iK|.

-

-

-

-

-

-

-

-

-

-

—

Avc.umc doiUi rclaHvc lucrciiienl' I'l sr..ni liciqlit-.

■47

-rs

lOO

cj'

9'

91

t03

^-

C.3

-

5-3.

ao no

Avcroqc dc.^ rcluttve lacrcnicnt- in

70

qs.

1Z5

TS

98

173

S09

■ 2ti

&<J>

-

^G

Avcrouc dail-.^ rcloKvc tucrcntcut" if? drv) wt;iLjt7t.

ec

no

)4o

qs-

119

172

'9'!

132

B-A

-

.,

ll')

BALTIMORE!

JUH£

I MAY jurtE iS

JULY 9

23

Aua

as

5EPT 3

AU«. 5EfT LO 3 StPT^OC T. 1. 1 .

SEJ-E 19 OCT. 11

ccr. 1 OCT. 3"

/NV.

Colt-urG numbGr

'

a

3

^f

.5

to

7

e

9

JO

(1

Lc>TCi+-li of qrovM.nq per.oO, Oa^^s.

Zl

£T

2.°l

za

se

se

se

30

20

zs

30

Number of plaaVs.

-4

S

s

e.

to

<t

b

S G

.5

3

fCcmu'iidci- sui-nniQ+ion I'ndGx.

82r

678

997

•OiS

"os:

10.^

lO-li

597

&05

o20

«I9

ae3

tcmpcrcjtura. inde*.

OS I 1 11

12^

mi

151

\30

I50

lOI

e.

o5

At

1 lO

Averaqc dail^j mean fempenatur-e^deq F

70

-^ 7 J

77

77

7T

o9

(^•4

^-<

GO

71

Avcaqe dail\j relative. i^vaporoTion

MS

t IZ

c»3

Se>

i02

,,,

qe

7J

<39

oe

e.1

qo

Ai.«rai:^e dail^j relo+ive aunsbine i'n+en5i1^-

1(5

q9

QA

eo

at>

7C>

77

7S

eti

-i-n

So

az

lOO

I 19

13a

IZZ

122

131

122

91

G9

IS

J&

104

90

lOq

IIJ-

W7

13^

111

19G

37

72

79

lOl

11.5

Avcraqe dQll^^ rclo+ivG iiTcremenf If? dr-\i VVCIC\ ht.

oe>

1 1 J

iiq

1-1&

1 JJ

Me

,..

92

-.■o

o7

1

i05

DARLin&TOM

WAY 1.5 JUNE 13

MAT 30

JUME 13 JUl-Y

JUME JULY £-1

JULI 7

AUG 21

ALl<i 7 SEFr

Al-Ki SEPT ■a

5EFT OCT 2

;?En >e OCT

Av.

Culture numbciT

1

2

3

H

5

O

7

6

-J

lO

Lenqth o| qrowniq period, Joms.

£9

a?

^7

ae

2S

se

23

ea

2e

2©

Number of plonfe.

G

£

&

A

-S

5

-

3

&

S

Kcmoinder summation Index.

Ooi

ez-1

oa-1

9^H

11,8

7t>2

7TD

lai

t.2.)

»>53

692

Ai.GraqG. dailH relai-ivc |^l^nslolo^ -oal tempcra+Lii^c mdcx

93

97

91=

ie4

I3<

I30

13-S

35

^3

57

lOO

AvGraqc dQll^3 iTjGCin "terrjpcrature^dc^ F

6S

70

70

73

71

T^q

75-

S7

fc2

■3.3

70

Averaqe da»l\i relofve evaporation indcv

i£e

toe

T6

ca

76

70

7i

73

aa

aq

OS

Average do'i--} rialati've sun =■ h me

I'^i

111

S5

91

,„

1 1^

<57

a3

91

70

lOI

Avcrjc^c da.iv, relative incrc nient" ni

103

100

iiG

>n

n2

i3-»

-

79

-»4

03

lO^

Avcrjqc da.W relative: » iicrc mei it n,' lea J" area

l^ie

ife".

iSl

iqo

^ >o

1<^3

-

90

.5 7

SH

wa

Average doilvf <T3loti vc mcreiTiGRt ui drvj wciqtit.

I^H

.-le

no

i7e

....

>&'■)

-

69

5o

72

„.

PLATE V,

118

COLtMAM Exposed st/stion.

13 JOME

M^V JUNE ZA e

JUUT e 5

JULY Z2 AUG, 19

5 1=) SEP7 sen 2 IG

3EFr 2 30

5EPT IG OCT. 13

5i;n 30 OCT. ze

Av.

Cul+ure n«-n-'itaen

I

C

3

1

5

6

T

e

9

lO

11

Leii<^i"h of qrosNioq period, davj5.

Z<)

SI

27

£e

Z&

28

Z&

2e

se

2t

ae

MLunbe.- ol plu.tt-s.

G

G

^

■^

&

e

.s

G

.s

■1

3

Kcmaintlei- sumoiofion I'ndcx.

070

«'J9

,..,

io4e

OT7

053

lose

9IO

709

-

-

959

Average daiUj relative phxjsioloqical teiupGixiture index.

9T

it«3

iZA

I50

iss

15E

153

.,<,

83

-

-

isa

Avcraqe dail^ i-nean ^mpGrort'ure,deq. F

Gq

TO

Tl

T7

76

77

77

72.

67

-

-

73

Avcraqc da'ilM relative GVa poral'iot-i mdex.

M&

t-lS

l£.Q

isa

M'^

137

lOO

1 1^

131

IIG

95

I2B

Aweraqe daily rclotiwe suiishuie ..,+eus.t>,.

1.51

M3

93

tOI

.2.3

IZI

lOZ

-

-

-

-

12.1

Avcraqe do.iu relative mcrenieat .a

69

ei

9-1

lOG

,co

IOC)

a^

12

5-3

50

53

ao

-^vcrac^c JQiW relative inoreinent lil leof area.

ll<5

ISO

IGl

lAA

136

14S

loe

e9

71

-

17

in

Av^ rac^? doilN) relative. i.icreniGnt in

ICG

.53

159

*A3

131

116

85

1 '5

t.7

75

13

113

,

EASTOrs -^-u.'CGK per.od5

e JUME S

1-1 V 2Z

a

junt ^.^ JULV 20

AUli. 3

dUL-V so AUI^ 17

3 31

Aua 17 AUa 11

AUG 31 5EPI

sen ocr.

SEPT OCT. ZG

OCT. li nov.

AV.

Culture xTwinbcr

'

2

J

1

5

(=>

7

S

9

■o

II

i£

Lcnqth of qrovvii?q period, doi^s

3\

ae

ZB

2a

es

ae

sa

2&

20

^7

2e

ae

'Hii.tJhjcr oj plonhs.

£>

4

3

&

1

5

3

S

s

e

G

3

Rciudiiider summati'oa indCK.

eo9

TO^

913

,oa4

iO<;7

looa

1032.

911

I ti

707

G73

M99

6o6

Avtjraqt* da.l%^ relative plupioioqicol temperature. >ntie.y..

9a

„,

lai

1-13

m

139

IHJ

117

aa

70

3G

39

|05

Avcrucje dail^J iTKian te.mpcrafure,dcq F

Ga

Tl

-73

Tli

76

73-

,..

72,

a7

G.S-

»3

3a

70

Avcraqe duili^ relative cva porution index

• o

isa

ll£

e--)

HO

131

lai

112

103

9G

33"

93

109

Avcrutac ciu.U relative sunsOine intensity

159

I5T

lit

IZZ

IIG

113

lOS

90

9G

ao

SB

60

loa

Avci^,»HC dijil« relative ii->cremcn1" in s^cm heiqljr

17

7Z

91

lOJ

1*33

l>3

lOJ

ai

3"G

so

36

JO

73

Avcraqe Juil>^ relative ineremenf" in leaf ^rsa.

7-1

To

1 II

qo

lOi

lai

i£.i

9Jr

G(S

G1

3e

e

S£

A\cro>]c dail'.j relative increment ii7 drq i^epqht.

eo

Ol

n^

91

Hi

iZ.1

12.1

til

7Z

■la

32

22

35

done

JUnE Z3

s JULl

JUfll £1

AUG

AOG

seiT

re 5trT 15

SfcFT 20

15 ocr. 12

SEPT OCT Z7

Av.

Culture i/um ben

'

2

3

1

S

b

7

s

9

lO

n

Lenqth a^ qro^n/q period, do\fij.

^e

20

£9

•sa

sa

SO

£■3

es

2tt

27

2tt

r^uiiiber ot- plants.

ti

S

5

o

(3

o

3

<3

t=>

-5

&

KCiiioiiKlcr SLiiniTiQtion iudck- |' 7-- >

104

T )T

I00 7

■nT

991=

I03J

">OT

75C

u le

(aM3

eso

A\/erac.K: da.Uj relative. pKijsioloqtcol tenif^rof^re 'inde*..

til

IU2

m7

OO

I30

130

113

no

76

Cfe

52

ICG

Averaqv. dail\^ I'neon tenrjperQtMrG.Jeq- F

61

7J

IS

7J

73

73

7G

7e

06

0-4

G2

7 1

A*eroqc daily rela1"ive e vafJoratto»7 inOfcx.

l£G

-

-

-

73

92

90

ra

78

71

J3-

83

Avcraqe doil^ reio1-ive sunshine intciiVit-j

11 J

Ml

91

ar

9£

BO

ai

aj

ei

73"

5 5-

e7

Averuqe dally reJative incrcmen r iw stem heTql(+.

6>

ee

-o.

,oo

i3e

125

15©

t"i

(bj

5to

.»

92

Avcraqe doiU nelative MTerement tn leof oreu.

I2.S

155

no

-

l^7

132

206

.02

,,

7Z.

3o

117

Avcraqe dail\( relotive increment' n' drij wciqljt:

r...:

,,o

,3,

M„

13 /

li.^

-r,

'7 t

a7

»jr

30

M&

PLATE VI

114

OAhLA/HU ^■week periods. COVPRBD 3T/x-r(ON

^^ JUNE ■i

JUHt JUMP la

junr la JULY Z

JULY 13

lb JULY 30

JULl AUG. 13

AU'j AUQ. £6

AUl,. 51 fl sr.pr5EFr 11 t:4

AV

Cli Iture numben

I

£.

-5

^

3

fo

7

&

9

Lenqfli of qrowinq period, davj^.

13

M

H

i^

1^

1-4

t2>

lO

13

'HumDer of plaafs.

S

*o

&

5

«o

-

5

^

3-

Avcruqe cJaiKj i-clativc evaporation index.

nE

1*17

133

I09

m

iia

I IE.

as

9i

l£0

Avcraqc duihj rclativG i ncrcn'ieai"

S6

112

MO

ir,,

ISlb

-

e^

TO

5e

9a

Avcruqe dailM ralutive. iiicrcn iciit^ lit leaf -prod uot:

16

S5

1Z.1

nfo

15-1

-

09

1

q

7e

OAKLAHD 4-vM'eek periods. coveR&D 5TATlOr^.

JUME IB

JUME JULY

JUME la 15

JULY 2 JULY 30

JULY 15 AUG, 11

JULY 30 AUCj Zfo

AUG,. 5EPT

AUCi. ?r-, £4

1

AV.

Culture nu niber

1

2

J>

4

5

O

7

8

LeiTqth of qrowniq perfod, dovjs.

Zl

Z&

£T

26

aq

2.7

zs

2.9

r-*.urnljer o^^ plants.

5

6

■^

5

e

-

5

S

AveraQG daiKi relalivG eva poiahfon inde^.

lOO

\^o

1^1

1 13

1 li

* 112

99

79

1 n

AueraqG doihi relative incrGi-nGnl" in steni hciqlic.

5Cj

SI

OS

\OG>

97

-

63

6e

1 16

Averaqe dai^^J relative increment- in. leaf at-aa.

-

i07

\^s

>S5

lie

-

-

6&

Aycraqe daiKj rclat'ive incrcnTGn+" in di-q vMciqht:

51

qf

1S.1

12i

,n

-

SG

59

93

BALTIMORE ?-WGeK pGf-iOOS- COvEREO StAFIOM

junE io JUNE ^5

June JULY . 9

JULY 9 JULY e3

JULY 23 AUG.

AUG 6 AU(i- 20

AUCi 3

SEPr 3 sEPr 19

SEIFT OCT 19 I OCT. OCT. 1 , 14

AV.

Culture numbei^

3

-^

s-

e.

T

a

9

10

"

LcnqUi of qrotvinq pcfod, do\j3.

IS

M

lA

1-q

m

H

It.

• z.

13

IHuinber of plonts.

e

6

•4

6.

5

s

3

•s

■a

Averaqe cloil"^ relative evaporation index.

119

9-5

iia

109

11 7

Br

ei

TT

73

97

Averaqe dail\j relative incremGnt !n stem heiqht"

157

IfoO

Z.Zfa

£33

197

191

5&

TO

IS5

17^

Averaqe doilsj relativo increment in Icof - product

1

ISO

e&E

M7

leo

IG.3

51

2.1

^3

125

BALTIMORE q-weck periods. COVERED 3TATIOM.

MAY JUNE 25

JUitE 10 JULY 9

jurtE 25 JULY 23

JULl AU<i 0

JULY 2i

Aua 5CPT 3

AUC^ 20 5Epr 19

5EPT 3 OCT

5EPt 19 OCT 11

OLT OCT 31

AV.

Culture nuiTiber

2.

3

A

^

<3

7

S

9

10

1 1

Lenc^tli of qrow'mq period, da\j-5.

2.7

29

za

zs

ze

as

30

z©

as

30

Number of plants.

■S

-

t>

-4

fa

^

3r

3

-5-

■M

Averaqe dQiUi relative evaporotron indeK.

119

I07

t07

1 1 1

1 13

lOZ

64

79

75^

7 J

■)!

Ayercqe doiKi relotive "increnient* in t.teni lieiqht.

ISO

\ZB

13-1

231

1&9

\3i

131

-fP-

103

7Q

Ii&

AvGroqe dQiUi relative increment" in leaf are.a.

-

IBS

^.io

.^,

r/H

iJH

?1

|0|

-

1£4

AveroqG daily rcitatlve incremeiit- m dr^ wciqUr.

B1

S-^

1 10

lea

I03

13^

114

&/|

GO

9S

platl; VI 1.

115

i

EASTOrx £-wGeK pe.-iods. covEPeeo 3rATior*.

MAY Z5 JUME 8

JUME a JuriE £2

JUME ^^ JULY

JULY JULY 20

JULY ^o AUG

AUG 3 AUCi n

Aua 17 AUG, 3r

AUQ SEfr 31 l-T 5EPT SEPT

stPr OCT 1 1

A v.

Cu!+ure numbs r

E

3

1

3

e

7

©

q

lO

1 1

Lenq+li op qrowinq period, da\|3.

H

14

1-1

14

14

14

14

14

14

1-1

fSunibcr of plants.

4

b

^

3

e

&

JF

e

t>

e.

Averaqe doilxj relative cvoporation \n.dcx

166

135

SO

1 12

144

IE9

131

131

i04

lOS

1Z4

-Avcraqc doil^ nGlati\/e mcremcnl'in sfeni hG'iqh.T".

IZ'^

1 la

155

n 7

n 7

117

ie5-

OG

TO

-

\^S

Ayeraqe dai/\j i-ela+ive i noremeiit' ia leaf- produci+.

.^z

....

eq

leo

les

ElO

225

IZ5

55

-

\^Z.

'

1

&A5TOM -)-week periods. COVERED STAriOM,

MAY 25 JUNE

JUPIE Q JUL'I

JULY 20

3

JUL^ AUG iT

AUG 3 AUG 31

AUG. 17 SEPr 14

AUCi 51 SEPT

5EP1 SEPT OCT 14 ZS> 11 OCT OCT MOV

AV.

Culture nonibcr

s

3

■4

3

6

T

O

9

lO

\ 1

12.

Lenqth o|- qro^ivnicj period, daus.

z-a

ze

as

ae

^S

ae

ae

2a

£T

as

afe

Number of plon^-3

4

<a

&

-

&

G

.5"

&

t>

-

-

Avcraqe dodsj relative evoporotioa index.

i5i

\o&

qc

126

131

13a

13'

1 la

lOu

1^^

« I.e.

I2.0

Averaqe cioii\j relative inoremei-^t m stem lieiqlTt.

103

\Z.&

.t.5

-

131

I Sis

125

73

ai

-

-

U6

Averaqe doi 1 \j relative, mci'et'nen.'i" ia leaf area.

1 n

191

SO

-

13©

^o^

169

I04

es

-

-

157

Avcraqe cioiKj rclo + ive incren-tanf i VI dri.j ^VG.iaK.'t.

loa

I1-1

ae

-

lOT

140

1£9

I05

56

-

-

lOt^

BALTinORE FOREST 3TATiOrH.

junE JULY

JUL( q JULY 23

2i AUC^

AUGi Autb

Auq 20 3EP1 3

5CFr

5EFT OCT

OCT. OCT 14

AV.

Ct-il+Lire number

4

5

&

7

e

9

lO

1 1

Lcnqth o(^ qrowinq period, do \j5.

lO

11

14

14

14

te

12-

13

Muniber o( pla.its.

6

J^

-

^

-s-

'^

e

3

Ayeraqe dail\^ relative evaporation mde-x.

-

T5

SZ

e.7

7£

64

57

SZ

CT

AyeroQc doiKj relQ-tive J ncrenieal- in et4m heiqht.

2&I

456

-

i96

2L&4

n(

121

ZO&

ati

Averaqe daiKi relatvvG increment in leaf- jz>roduot.

9^1

I04

-

13(>

5q

13

G.

£1

sa

BALTIMORE ■^-weeK pGi^i'odS). FOfiLEi-T 3 1 ATlOrS

21 JULY i5

JULY AUQ. &

23 AUO. EO

AUli G SEPr 3

AUG. ZQ 5EPT t1

SEPT 3 OCT

5CPT iq OCT. 14

OCT, 31

Av

Culture nunfber

1

.S

t)

7

s

9

lO

II

Lcnqt'h o( qrovvinq period, daij^.

32.

£B

2e

^e

30

za

ts

30

Mumbcr of plants.

&

-S

-

4

S

J

6

3

Avcruqe daily relative Gvoporotfon inde><..

-

79

»- T5

70

ea

(3l

55

35

66

Averaqe <:iQi\s^ relative increni-enh in stem heiql^t"-

Z^o

441

-

310

£13

EOO

200

1^5

Z56

Averaqe doihj relative incremenV

lO^

I04

-

9-5

7i

10

ae

3?.

6C*

Avcr-oq^a UoiIm relative increme^it 1 11 rji- vj w''iql-(t.

•5^.

GZ.

-

4&

37

30

53

a?

4>

PLATS VIII

UG

platl: 1a.

Two-week graphs for Oakland, Chev/sville, i.^onrovia, College 3altir.ore, and. Princess Anne.

Ejctoosed stations.

117

e:x\>

O -J i~ I-.'

■^

w ^

^

•^	^>-, ■^	'■^		^	>		^
	1	>5			>- ^		< .0

Lb ^

11 .s

(lAV June junP Ji'ME J'.'!-!- .'•'-'i.y a.v- aucs. Seof

■i?r cv:^]-

.2'^

^^ 119

2- week per lod.

EXPOSED ST»*^T10tN.

riA> JUN£JUME w>i>Ne JULY JULY AUQ AUQ SEPT.-SEPT, OCT. i& ( . l>5 2=^ 13 2T (O S-H 7 ' ^1 6

^ &(3) ^j (£.; fe

EXPOSJEP STyVT\Oi^

1 **

i\?M

lo -

r-f/^Y June juhE Juuy juuy juuy aug, a:

BAuTrMOR£ ^yr'ose.o STATION

121

HAY C\AY JUNE JUNE JUuV JulV '^'OGt. M

l^^

9C-

70-

60 i-

i ^-

30-

\ \

«T)0

l^i^i

PLATE X.

Two-week grpphs for Drrlington, Coleinan, and I:Iaston. ?our- week, graphs for Cakiand, Chev.sville, and I-.onrovia.

Zxoosed stations.

m

40 h

I

I 20 1

\0

O

) '^- ^jU ^^' Np "-^ vp.' 4- -* -i'-

Ay<^ AUi^ 5E\? 5izP"0CT:'

7 2-14 la z

week periocJfc).

'^b

30-

*

10

__!_

V I \ \

\

MAV MAV JUNE JUNE JOi-Y JUCf AU^.Auq, SER ^£T* SE'P OCT-

120

121

^-\hJ^f^\'C periods

MAY J UH£ JUNE Jl>'U^Ji7L>< Ju^i Z'^, ■^..•^.■•^ . Z^ S \9 3 (6 3r iH ^

12H

Ex Pos-cU - . . . . ' orv

MAY JUNE vJUlVEj UN* JULY JULYAL

tT""

U) ® ^

^s3-' %>■• ' v-w

nu

4-week: pe^-'.ods.

May JuNEJu-ZEJUfys,

\

a

ISO

PLATK kl.

/our-v.eek graphs for College, Baltimore, Dprlingtcn, Coler.an, E?ston, ? no Princess Anne.'

]^xposeu stPtions.

131

20C — I30-

iSOr

iTDi-

I l60j—

i

1 5-0 1—

I

l3o|- \ZO -

I

i

lUC

90 i- ',

6o ~

- ;/ /

70

60 - /

■^.

-C. STATION

A-

nAV MAY JUHcJONC^-^, ,^— , w^,., .~.\JG -i'

zoo-

180-

n)t

170' 160I-

etiKpfc

Ii/.Y KAY JUKE JU1S J'-'i-X vj '-iw

LV^.'; ,".

1 ^

138

DAFfLinqrroN

-^-VV-e.^,^ per! od s

1 M

MAY /*i'*»M' Ju-

■5" zo i:;

i^ r^^6?^

JUUYv/ULY AUG Au(3,

lO 2^f 7 ' 72; ■

^4

lo*

MAY ^:5,.^' Jl'WSJUWE.JC'uX JULXAUC^. ALi(^.5£PT5E/5fSEfT

1 7-

135

(-TAV MAY JU^EjUfYEJ

iP (y

■' Vi/ v;-

136

U S<S a 23 7 2\ .*q. le I

'V

^,

137

PU^.TS Xg.

Two- and four-week graphs for the covered stations at Balti- ;cre and Oakland, and for the Beltiir.ore forest station.

IHH

CO Vn:\^EO STATION.

t-A AY J uKEJUNPOUCi JUp/ JULY AUQ./NUQ 5EPT "

139

OAKV-AMD COVef^EO STATIC r<.

^Z -«:| i© 2 ii^ 30 J3^ ZG

^^^ /'^^ {'"2^- ^-""P; f7^ (^>~\ fZ^ v>y' tsjj ^^rv k:;^ xj£x VwC.«' v.5_^

140

HI

\5ALXi\10\ZB.

\ j era V F_if £ p ;est,^-^.' o rx.

^— \ — IV-

C2:) cl> ® cS^ (D q>(S) ep do) ^

142

Uku-

"300—

t

I

l8o-

i

7^"

\

\

,2^

eo -

i

1

\

^ ® © ^ cB> (g

/I

^

lis

7^l\e.^T 5TATIOM

T.

^ > ® f ^.^ 6) (S^ <3> (li^ vLP

v:?.

144

PLATE XIII

Two- end four-week graphs for the covered station et l-i&stcn. Graphs of seasonal averages for all stations. Explanation of conventions usea in representing the plant and climatic cuan- t it ies .

145

£A3TOK 2.- vvee-lc p e, r / c?d <=)

SO

' I I I

146

cove

4-week; p-

© © @ (© c£) C!) C^ @ ®

5\ H ■

X^/.

1		vjO
o	Oi	X UJ
in

^T

\

V A

|0

I- ^'

tlJ

"/

y

.^C-

)i'

iti

4

0 11

■\\

t-

irv 'r\

^

JL L

^ 9

-i__ \ V-

o

0

o

'0

148

(T) >T ro >-\!

Legend.

.Explanation of conventions used in the representation of the plant and climatic ouantities.

Dry weight.

Leaf-product,

Stem height.

t^^-^r^t- :<^!**>.t*:<>«*<*'*M,JlcMrmm>»rM«MM>tiiMeD«ilUB.MM««*i«ini»*»»'ww

.-jw »'jt»t'»r ^wii*»<rTw^^w .lai

Temperature index values .-. Evaporation index. ~^ ™- Sunshine intensity.

The dates shown are those of the first day of each culture period! The numbers under the dates are culture nunibers corres- ponding to the numbers' shown in the second lire of the tables. The numbers in the vertical column at the left give ordinate values.

15(1

VITA.

The writer was born December 26, 1888 at Baltimore, Maryland. He entered the Baltimore Polytechnic In- stitute in 1903, graduating in 1907. During the year 1908-1909 he taught in the public schools of Baltimore. In 1909, he entered the CJollegiate De- partment of the Johns Hopkins University, receiv- ing the degree of Bachelor of Arts in June, 1913. During the years 1914-191'( , he attended the Johns Hopkins University as a graduate student in Plant Physiology, Physical Chemistry, and Botany. He was engaged in research for the Maryland State Weather Service during the year 1915-1915, and carried on research for the U. S. Forest Service at the Utah Forest Service Sxperiment Station dur- ing the summer of 191c.

a^

"i^^

-K.

'i^W^^^%^^

s.^'•A■■

"ml

!,,■>■■■!-:'; ^v:v;M/,vtv:s'A.>7i;^riw,fi>-

^OiV^i^OCMX

^MMMMm.

■J%A

{ : •: >i )! w ]Qhi'd. : ' ■ >i^^.. .... .-. .... ....._

, a .n 4 4^ ii. Ri »i v^.M ^

[ ^i '^ H M m\^ „ . ^ : ^f -^ -' ■ . ., :, ■-- .T ...

Aj4 M /^ ^X t4 ;4.M .H K K A ^ '^ ' "

m, f4 nH n . : ii Mi* i^ ?€ M^i( '^ H 'u n n n n >

... U H H

M 'i*t' Mm M '^' '^ '^ M i^ n n''^ '

,K.jN:J4i^^ M..M>r

^ St. ,U M ■

m %■ V

, : jl / m !^l I'* i^ n. A ^ ^ 1^ n ^ i^ i^ i^ ¥ . , .^ It A. ^^^ -^ -^^ '^ -^ n ;i 'If m \i. 14 \iii V ^- -^

^ >1. /I /i ^-^

I yi ytjly

i M MM >J^>0^ H\. V n -^i, )% U 9ii li ti ii .^ hi u

V«^ )*- '.^ I

^ ^*^ 'i ^i A *1 ^'1 ''^ ^ • ■

:% n- A- *'*• ?A

*'. *^ U f.l */ if itt »*■

iff ^lisii-'f-i

tV-' ^tJ ^RJ

I

^ ^ In h

^K-X-CW

«>:x ,. ^V*.^-"- ■»- *•>»■■»"*

lA"

Ti

i

M n M v| ?;| Ri e$ i^ H V

».-t ..f «:u .£ .' vW ■ .•«■

iW »l M l«f ,1/4 V.*' ivt; ,>i iif i/i its. ,A i'i '(.».* \'i- W "i'^' .*.. 't.

«

llin

Til

.,.'■. .^^ ,v\ h,

H

^^- ^^^ J^ «M i<^ :*^ -^^ ;^ >^ *^^ >t ^ ^4 ?^ 'H ■^; rf m -^i

. ii M M ^ H ^ ^t »i H rC «^

t r4 :^»^' »^( ^ ■■( -n* \f ^» h '«

' . M H -fi H M >{| M M ?(

. ,^ ... ^ ;. . ., ?< ■ M; «^( ' V - •

i^.jv^. r:i>4 nji ^' ^- ■■■

dM

WW

*l ■' *

14 H H

.'Ji M'u'*;

?,. ;4. .*.l. ^'

d rl n )i.. )i \i

■■ n >n H m ■■ ' ' '■'

4 1.1 n U r4 M .^ t

i! >J»-.

i!f vi~ m' '"^^amX

'.*. Vi

'.I

J4 H ;t j^^j^i

m^

■« : u : i M ' ' f-i H -'x '

\M&9m

A: «i i««- j4 jh '^1

■

it.|>i I