UNIVERSITY OF CALIFORNIA • COLLEGE OF AGRICULTURE 
 
 AGRICULTURAL EXPERIMENT STATION 
 
 BERKELEY, CALIFORNIA 
 
 PLANT SUCCESSION 
 
 ON BURNED CHAPARRAL LANDS IN 
 
 NORTHERN CALIFORNIA 
 
 ARTHUR W. SAMPSON 
 
 A CHAMISE BUSH, AND THE SAME PLANT ONE, TWO, AND FOUR YEARS AFTER BURNING 
 
 BULLETIN 685 
 MARCH, 1944 
 
 UNIVERSITY OF CALIFORNIA • BERKELEY, CALIFORNIA 
 
CONTENTS 
 
 PAGE 
 
 Foreword 3 
 
 Uses of chaparral lands 5 
 
 Ecological characteristics of the chaparral association 8 
 
 Growth habits of the chaparral 9 
 
 Climate and soils of the chaparral 11 
 
 Special characteristics of the chaparral regions 13 
 
 Survey of viewpoints and practices concerning brush burning 18 
 
 Burning by Indians in California 18 
 
 History of state fire-control policies pertaining to brushlands 21 
 
 Experience and attitude of stockmen concerning brush burning 23 
 
 Plant-succession studies on chaparral lands 26 
 
 Review of results by earlier workers 26 
 
 Organization of the study 28 
 
 Herbaceous plants common to chaparral areas 29 
 
 Eesults of plant-succession study 30 
 
 Recovery of brush after burning 31 
 
 Invasion of herbaceous plants 40 
 
 Widening of chaparral areas by burning 56 
 
 Replacement of chaparral by grass through exclusion of fire 59 
 
 Soil temperature during burning 61 
 
 Heat studies with seeds 64 
 
 Reseeding of burned areas 67 
 
 Effect of chaparral ash on seed germination and growth of grasses 72 
 
 Soil-nutrition and plant-growth studies 81 
 
 Background for the studies 81 
 
 Results of chemical soil studies 83 
 
 Chemical composition of vegetation on burned and unburned areas 90 
 
 Field observations of areas burned under state permit ' 98 
 
 Methods of brushland improvement , 108 
 
 Removal of brush and tree growth by methods other than broadcast burning Ill 
 
 Comparison of brush-removal methods 119 
 
 Comparative grazing values of different brush covers 120 
 
 Effect of fire on wildlife and recreation 124 
 
 Magnitude of the brush-burning problem 127 
 
 Summary and conclusions 128 
 
 Common and scientific names of plants mentioned in the bulletin 135 
 
 Acknowledgments 138 
 
 Literature cited 139 
 
FOREWORD 
 
 The foothill lands of California have long been relied upon to furnish pas- 
 turage for large numbers of cattle and sheep. From the earliest days of settle- 
 ment stockmen have planned their yearlong operations in accordance with 
 the availability of forage in the different climatic and elevational areas. The 
 most common practice is to graze the ranges at low elevation, nearest the ranch 
 headquarters, as soon as the forage growth develops in the spring, and to hold 
 the animals on these lands until growth is properly advanced on the summer 
 ranges at higher elevations. 
 
 Although the greater part of the foothills support grass and other herbs, 
 this cover is interspersed with brushland almost throughout the length of 
 the state. This "hard brush" cover, characterized by simple, thick, leathery- 
 textured leaves, collectively called chaparral, is composed of many species; 
 but all have in common a dwarfed habit, an extensive root system, and the 
 capacity to endure long, dry summers. As a whole, this chaparral cover in- 
 habits thin soils of low productivity. The purer stands of the chaparral asso- 
 ciation occupy some 7,300,000 acres, or about 7 per cent of the total area of the 
 state, and clothe the lower slopes of the mountain ranges which line the coast 
 and flank the great valley areas. In addition, there is a large acreage of wood- 
 land and of cutover timber areas which have been invaded by a mixture of 
 chaparral and thin-leaved brush species. In general, these wooded areas have 
 relatively deep and productive soils. The brush-fields, like most of the uncul- 
 tivated lands of the state, have multiple uses. Extending, as they do, into the 
 timber belt of their more elevated northern distribution, and into the coastal 
 lowlands in the southern part of the state, they are of importance in the 
 production of livestock, game, timber, and as protection of watersheds. 
 
 The chaparral areas furnish limited seasonal pasturage to supplement that 
 of the grassland of the foothills and the forested lands of the mountains. At 
 best, however, the chaparral cover compares poorly with the grasslands as a 
 source of forage, since the brush itself is of low palatability, and the under- 
 story herbaceous vegetation is sparse and largely inaccessible. Thus, the ques- 
 tion arises as to how the chaparral areas may be made more permanently 
 productive, and how chaparral invasions on the wooded lands may be curtailed 
 or controlled. Opinions differ as to whether these brush invasions are largely 
 the result of overgrazing and burning, or whether they are partly due to 
 fire suppression resulting from restrictive state fire laws. Some believe that 
 burning by prehistoric Indians was the natural means of suppressing the 
 brush on these lands and that controlled burning today is the most, if not the 
 only, effective means of increasing the yield and of improving the quality of 
 the forage produced thereon. Stockmen who own brushlands, and who must 
 rely upon them for part of their livelihood, naturally desire to obtain the best 
 returns possible from them. In attempting to improve grazing on these lands, 
 the most common practice is indiscriminate ("broadcast") burning of the 
 brush. Although this practice may not result in converting the brush to grass- 
 land, it does open up the brush cover and temporarily favor growth of browse 
 and of some palatable grasses, the volume of which depends largely upon the 
 productivity of the soil. 
 
 [3] 
 
4 University of California — Experiment Station 
 
 Because various administrators and land-use investigators believe that ex- 
 tensive brush burning may not bring economic returns, and may jeopardize 
 certain long-term public interests, controversy has arisen as to the best meth- 
 ods of management of the brushland areas. On the one hand, the stockmen 
 who must pay taxes on their brushlands and earn a part of their living from 
 them, feel justified in employing any rational method which promises even 
 temporary increases in pasturage. On the other hand, public-land admin- 
 istrators contend that reasonable restriction in brush burning is necessary 
 to preserve important watersheds and various other values. These differences 
 of opinion are indicative of the complications of the problem, as well as of the 
 lack of understanding of the issue as a whole. They are due chiefly to the fact 
 that conclusions have been drawn largely from mere observation of practices 
 instead of from the accumulation of measured results of scientific investiga- 
 tion : For many years, biased points of view for or against burning and even 
 propaganda for some preconceived philosophy, in the written and spoken 
 word, have confused, rather than clarified, the issues involved. 
 
 A broad and comprehensive understanding of the problems involved in 
 brush control requires research in a number of sciences, including agrostology, 
 silviculture, botany, geophysics, hydrology, soil physics, animal nutrition, and 
 economics, as well as knowledge of many administrative practices, involving 
 public and private land use, wild life and forest management, fire prevention 
 and control, forage production, water conservation, and soil-erosion control. 
 
 For this reason I have encouraged especially qualified investigators of 
 the College of Agriculture to study carefully the methods and results of brush 
 burning over a number of years. In order to assure the broadest possible 
 approach to these studies, I set up in the College of Agriculture in 1933 a 
 standing Committee on Range Management, consisting of nine investigators 
 with special knowledge in the fields of agricultural economics, agronomy, 
 animal husbandry, irrigation engineering, forestry, soil technology, and zool- 
 ogy. Among the studies which the Committee has already undertaken are some 
 dealing with the effect of brush burning on runoff and soil erosion, silting of 
 streams and reservoirs, water penetration and water-holding capacity of soils, 
 and the ultimate vegetative succession. The findings of these studies will be 
 reported by the College of Agriculture from time to time as evidence is 
 accumulated. 
 
 The present report, based upon studies conducted in the northern part of 
 the state during the past fifteen years, presents essentially measurements of 
 changes in the plant cover on burned brushlands, introduced by a brief his- 
 torical resume, and with some discussion of related points. Some members of 
 the Committee feel that the report presents too specialized a viewpoint of the 
 problem. It should be noted, however, that the bulletin does not purport to 
 answer all questions pertaining to brush burning. Nevertheless the data pre- 
 sented should be helpful to those who desire information concerning plant 
 succession and related subjects on burned brushlands of the types studied. 
 
 C. B. Hutchison 
 
 Dean of the College of Agriculture 
 and Director of the Agricultural 
 Experiment Station 
 
PLANT SUCCESSION ON BURNED CHAPARRAL 
 LANDS IN NORTHERN CALIFORNIA 2 
 
 ABTHUB W. SAMPSON 3 
 
 The chaparral lands of California must be regarded as having an integral 
 part in any rational, long-time land-use program. In the formulation of such 
 a program the more productive areas may prove to be of greatest usefulness 
 under private ownership and management. On the other hand much of the 
 
 
 
 
 
 
 
 IP 
 
 
 
 ' r -^ : - - ^ /i- \ ♦ :-' 
 
 
 ■ffifc* 
 
 
 
 ' " ■ ■ - . ' .. ... 
 
 f Si i 
 
 
 X ^ :; fef ■ < : V" , f*iJi8$fc 
 
 " '^- .. ; 
 
 < 
 
 Fig. 1. — Subsistence homesteads are often occupied by industrial laborers 
 who can obtain only seasonal employment. Chamise area cleared for growing 
 grain and orchard crops, in Mendocino County. 
 
 less fertile chaparral lands may prove to be most valuable in such public uses 
 as watershed protection, recreation, and game production. The analysis which 
 follows points out the dominant industries and uses of the five major chaparral 
 regions of the state. 
 
 USES OF CHAPARRAL LANDS 
 
 The chaparral areas extending along the dry eastern slopes of the north- 
 ern coast ranges, from Shasta to Napa counties, are of importance both to 
 private owners and to the public. Where these stands are not too dense, or 
 where they give way to grass, they are used for spring and early summer 
 grazing; also they are of great importance as watersheds in some localities. 
 Moreover, the large resident deer population on these lands constitutes a 
 
 1 Received for publication October 17, 1942. 
 
 a Acknowledgment is made to the Works Progress Administration for assistance rendered 
 through its Official Project No. 465-03-3-630, Unit B-12. Grants for the support of this 
 project were also obtained from a special state appropriation for research in forestry in 
 cooperation with federal agencies. 
 
 8 Professor of Forestry, and Plant Ecologist in the Experiment Station. 
 
 [5] 
 
6 University of California — Experiment Station 
 
 valuable hunting 1 and recreational resource accessible to large centers of 
 population. 
 
 The western half of the northern coastal ranges, extending north and west 
 from Santa Rosa and St. Helena, in Napa County, to a point 50 miles or more 
 from Dyerville, in Humboldt County, ranks among the highest of the chaparral 
 areas in economic importance. The slopes adjoining the narrow, fertile valleys 
 are covered with chaparral and woodland-grassland, which are used exten- 
 sively for the growing of cultivated crops, notably, fruits, grapes, and grains 
 
 (flg.i). 
 
 Timber production is also of importance commercially. A large island of 
 second-growth pine, which is readily accessible and therefore of rather high 
 potential value, is found in Sonoma County (fig. 2). The all-important red- 
 
 '/zxM, Second growth ponderosa pine 
 Redwood (virgin) 
 Douglas-fir 
 
 Fig. 2. — Distribution of timber regions in the foothill areas. 
 
 woods also occur in this region. These trees produce high yields per acre, and 
 the cutover areas are readily restocked by sprouting or planting ; this favors 
 a permanent forestry industry {74)*. Except for the highly fertile flats, at- 
 tempts to convert the redwood cutover lands to agriculture or pasture have 
 been largely unsuccessful. These lands are difficult to clear, the invading 
 plants are of low forage value, and artificial seeding to cultivated grasses is 
 seldom successful (17). The lands around Clear Lake and along the Russian 
 River are of high recreational value, being used for hunting, and in the more 
 favored areas, for homesites. Lake, Mendocino, Humboldt, Sonoma, and Marin 
 counties head the state in numbers of deer killed per square mile ; but other 
 4 Italic numbers in parentheses refer to "Literature Cited" at the end of this bulletin. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 7 
 
 game animals are little hunted {45). These lands also function as water- 
 shed (81). 
 
 The relatively dry eastern-slope foothill region of the southern coastal 
 ranges lying adjacent to the San Joaquin Valley, south from San Francisco, 
 is predominantly covered with grassland and chaparral. Grazing is almost 
 the sole industry, large areas being relied upon for seasonal pasturage. Al- 
 though losing its nutritive qualities early in the season, the vegetation is so 
 extensive and diversified as to afford fair-to-good grazing for various periods 
 of time. The hunting of deer, pigeons, quail, and cottontail rabbits is another 
 important land use. 
 
 The western portion of the south coastal strip has many uses. Recreational 
 activities in the Santa Cruz Mountains and adjacent areas, as far south as 
 Los Angeles, are enjoyed on a large scale the year round. Many week-end and 
 summer residential homesites are located in these areas. Small retreats are 
 especially abundant near the city of Los Angeles. The demand for this type 
 of homesite has been recognized by the federal government, as evidenced by 
 recent legislation. 5 The large populations of deer, cottontail rabbits, pigeons, 
 and quail afford recreation ; and coyote, fox, skunk, and raccoon, particularly 
 on the foothills away from the coast, furnish a fairly large revenue of fur. 
 From the standpoint of wildlife production the chaparral areas south from 
 San Francisco are the most important in the state. 
 
 The chaparral lands along the western slopes of the Sierra Nevada repre- 
 sent the most diversified agricultural uses of the units here designated, largely 
 because of diversity in climate, topography, and vegetation. The cover con- 
 sists chiefly of pine timber, oak woodland, chaparral, and grassland. Irriga- 
 tion favors the growing of fruits and various farm crops, but dry-land farming 
 and grazing are also extensively pursued. The livestock raised in this region 
 are grazed on the grass and brush areas during the spring and early summer 
 months, and a fair proportion of the animals are moved to the mountain 
 ranges of the national forests for most of the summer, being returned again 
 to the foothills in the fall. At the upper limit of the chaparral areas is an inter- 
 spersion of second-growth ponderosa pine (fig. 2). This forest area possesses 
 great possibilities as a source of future timber. Although the stands are some- 
 what understocked, many sites are exceedingly productive, and the annual 
 yields are high (16). These cutover pine stands are readily accessible and are 
 rapidly approaching maturity, so that they will presumably come into strong 
 demand in the near future. Their chief uses will be those of common lumber, 
 box wood, mine props, pilings, posts, shingles, and shakes (51). Some of the 
 lower areas which were formerly in forest might be replanted to timber, but 
 the present high cost of reforestation, the high fire risk, the present limited 
 demand for lumber, and the time required for the undertaking preclude 
 putting them into timber at the present time. The area is also used as water- 
 shed. Many species of wild life found here serve as an important and insepara- 
 ble part of the land values of this region ; they attract recreation seekers and 
 hunters from remote parts. Deer and mountain quail use the chaparral cover 
 
 5 On July 27, 1940, the Izac Five-Acre-Tract Law went into effect. Under this law tracts 
 of five acres are leased or sold by the federal government primarily for use as part-time 
 camping, health, convalescent, or recreational homesites. 
 
s 
 
 University of California — Experiment Station 
 
 as a seasonal foraging ground. Also many species of fur animals furnish added 
 local revenue. 
 
 With full recognition of the varied uses of the foothill area, the place of 
 the chaparral lands must be fully acknowledged in the development of a land- 
 use program for the state as a whole. Further study should aim at classifying 
 them into their major permanent uses. To accomplish this satisfactorily im- 
 plies intimate knowledge of the ecology of the chaparral cover. 
 
 Fig. 3. — Left, chamise seedling, one year old, showing taproot 10 inches long. 
 Eight, chamise seedling two years old, showing extensive root branching. 
 
 ECOLOGICAL CHARACTERISTICS OF THE CHAPARRAL 
 
 ASSOCIATION 6 
 
 Knowledge of the ecological peculiarities of the chaparral association is 
 basic to a consideration of the brush-control problem, and to the ultimate 
 management of these lands. The present discussion considers the growth habits 
 of chaparral vegetation, the climate and soils peculiar to these areas, and 
 discusses the ecological characteristics of the chaparral vegetation by regions. 
 
 e In the nomenclature and characteristics of plants discussed in this bulletin, the author 
 has in most instances followed W. L. Jepson, Manual of Flowering Plants of California, 
 Associated Students Store, Berkeley, Calif. 1925. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 GROWTH HABITS OF THE CHAPARRAL 
 
 The shrubs of the chaparral lands are closely branched, 2 to 10 feet in 
 height, and somewhat treelike in appearance. The leaves are mostly small, 
 thick and simple. Several prominent species have dual root systems, charac- 
 terized by widely spreading superficial branch roots, and more or less of a 
 taproot (fig. 3). The thin-leaved, deciduous, "soft brush" understory of the 
 coniferous forests of the upper mountain slopes intermingles extensively with 
 the more drought-enduring "hard" chaparral only in localities where condi- 
 tions of growth are rather favorable. 
 
 The chaparral species may be conveniently considered in two groups — 
 sprouting forms and nonsprouting forms; the most common species in each, 
 including some of only local importance, are listed below. A complete list of 
 species pertinent to this study, with scientific names, is given at the end of 
 the bulletin. 
 
 Sprouting species 
 Brewer oak 
 California buckeye 
 California scrub oak 
 Canyon live oak 
 Chamise 
 
 Chaparral coffeeberry 
 Chaparral whitethorn 
 Coast live oak 
 Coast whitethorn 
 Deerbrush 
 
 Dwarf canyon live oak 
 Dwarf interior live oak 
 Eastwood manzanita 
 Greenbark ceanothus 
 Indian manzanita 
 Interior live oak 
 Leather oak 
 Lemmon ceanothus 
 Mission-manzanita 
 Poison oak 
 Eibbonwood 
 Shagbark manzanita 
 Toyon 
 
 Western mountain-mahogany 
 Woollyleaf ceanothus 
 Woollyleaf manzanita 
 Yerba santa 
 
 Nonsprouting species 
 Bigberry manzanita 
 Bigpod ceanothus 
 Blue blossom 
 Common manzanita 
 Cupleaf ceanothus 
 Gregg ceanothus 
 Hairy ceanothus 
 Hoary manzanita 
 Jim brush 
 
 Mariposa manzanita 
 Parry ceanothus 
 Parry manzanita 
 Pecho mountain manzanita 
 Pointleaf manzanita 
 Bamona bush 
 Serpentine manzanita 
 Stanford manzanita 
 Stripeberry manzanita 
 Wartleaf ceanothus 
 Wartystem ceanothus 
 Wedgeleaf ceanothus 
 Whiteleaf manzanita 
 
 Stands of nonsprouting species are Rilled when heavily burned or when 
 chopped, but fields composed of sprouting species send up new shoots from 
 the crowns or rootstocks when burned or cut back. The root crowns of chamise, 
 like those of some sprouting species of manzanita, are distinctively swollen, 
 enlarge with age, and develop rapidly but irregularly after a fire. The pres- 
 ence or absence of sprouting brush species, as shown in later discussions, 
 greatly influences the methods and success of brushland clearing. 
 
 Another distinction between the different forms of chaparral is made ac- 
 cording to leaf width. On this basis are recognized the broad-leaved forms as 
 

 o 
 
 
 
 
 
 
 
 w 
 
 
 W 
 
 
 A 
 
 
 i 
 
 
 M 
 
 
 O 
 
 
 2| 
 
 
 M 
 
 
 O 
 
 
 w 
 
 
 > 
 
 
 E-h 
 
 rH 
 
 fe 
 
 
 o 
 
 H 
 
 
 h-l 
 
 S 
 
 pq 
 
 O 
 
 < 
 
 pel 
 
 H 
 
 M 
 
 
 c 
 
 JO GO O O «- 
 
 
 ■< o o >o »o 
 
 
 i 
 
 oo coco «o *o t(( « r~ i^ 
 
 .2 erf CO 
 
 
 
 
 
 
 
 
 
 
 
 
 £ o >> 
 
 61 
 
 < 
 
 oo cm 
 
 i O r- 
 
 
 ^ co i^ 
 
 lO t-- ^H ^ 
 
 
 (MM CM <N CN IN *N O 
 
 <N CM 
 
 
 
 
 
 
 
 
 03 
 
 in in oo m w oi *— li <t 
 
 
 IN CO lO O »0 CO r^. o co r~ 
 
 
 3 
 
 CO CO COCO COCO CM CO cm in 
 
 h 
 
 d 
 
 CO OS OO -* t— CO f <N CM 
 
 
 § 
 
 1 
 
 
 1 
 
 1 
 
 1 1 
 
 ■*- J £4 00 
 
 
 
 
 
 
 
 > 
 
 co o co ex 
 
 ) CN OS <-> CO CO CO 
 
 !D N l^ t^ 00 O 
 
 
 Oi i—i 
 
 — a) 
 
 ri « cu 
 
 »c«o iO«0 lO >0 »0 lO lO o 
 
 
 
 
 
 
 
 2 3m 
 
 2 "^-2 
 
 X 
 
 Op PP ^. o. ^ o_ ^ &. 
 
 03 
 
 CO CC 
 
 i co co oc - 
 
 ■1 CO CN 
 
 
 < 
 
 s 
 
 i-l CM i-H *- 
 
 r-H CSJ H M — < -H 
 
 .2 
 
 uuin^ny 
 
 CO oo lO 1- 
 
 CO c 
 
 5 1- CO f CO 
 
 
 
 
 
 
 
 3c 
 
 
 
 
 
 
 
 
 
 
 
 
 
 45 'C s 
 
 jauiturit" 
 
 CM C- 
 
 1 ^H -r-< COIN N N N IN 
 
 
 
 
 
 
 
 a 5 » 
 <3 &-C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 03 03 
 
 Suijclg 
 
 CM CO t^ Oi 
 
 IN M N IN COCO IN IN COCO 
 
 43 « 
 
 
 
 
 
 
 
 o 
 +3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 ja;ui^ 
 
 1^ t~ 
 
 lO lO lO W3 lO »0 to >C lO iO 
 
 
 d 
 
 
 
 
 O o o o 
 
 
 ""! °. 
 
 t^ t-- oo oo co co 
 
 en 
 
 i 
 
 CO IN 
 
 co co" 
 
 ep 
 
 CM CM 
 
 
 <N <N 
 
 
 
 
 
 
 
 
 
 
 
 
 (N lO lO >0 
 
 
 o >o 
 
 K N IN CO 
 
 
 «i 
 
 «* CO 
 
 d - 
 
 ; co X) oo f 
 
 
 CO « 
 
 co co co eft r-i co 
 
 
 
 
 
 o c 
 
 o o 
 
 02 
 
 
 t- "I- 
 
 
 «!)< C 
 
 UO CO CM CM 
 
 
 
 d d 
 
 
 — i oc co" CO 
 
 
 s 
 
 »o c 
 
 
 
 >o CO 
 
 
 
 
 
 
 — i CM CM 
 
 , 
 
 • 
 
 ■(S lO to w 
 
 lO CO Oi f -* CO 
 
 c 
 
 
 
 o c 
 
 pp — ; - 
 
 —i o 
 
 <5 
 
 do o c 
 
 d d 
 
 CD .2 gg 
 
 g -£ d 
 
 
 
 
 
 
 
 
 •Of f OC 
 
 — c 1- 
 
 co cr 
 
 CM Oi 
 
 343-5 
 
 
 p i> 
 
 
 
 -*l CM 
 
 3 a c 
 
 03 « rt 
 
 <j 
 
 o d o c 
 
 o <S <zi d do 
 
 
 
 
 
 
 
 <D 
 
 
 lO f 
 
 lO lO 
 
 
 03 
 
 t~ o co a 
 
 oo oc 
 
 T* Ttl t^. f 
 
 ~ 
 
 S 
 
 ^4 CN 
 
 o c 
 
 
 r-: ,- 
 
 O — 1 
 
 
 
 Oi — 
 
 O cr 
 
 oc oo O O Oi CO 
 
 ^_ 
 
 C 
 
 CO tc 
 
 CNJ cc 
 
 oq c\ 
 
 oo or 
 
 OO p 
 
 C 
 
 
 © o 
 
 »C c- 
 
 CO C" 
 
 
 CM IN 
 
 |e5 
 
 
 
 
 
 
 ^.2 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ©'-P © 
 
 
 IN W 
 
 uo -q- 
 
 — i cc f ir 
 
 co co 
 
 c -3-c 
 
 > 
 
 l>- CC 
 
 C3: OC 
 
 co cs 
 
 CM i— 
 
 -* 1 (M 
 
 Sac 
 
 <! 
 
 a- — 
 
 O ^ 
 
 co O uo C 
 
 d t~- 
 
 — c- 
 
 
 
 
 
 "" U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <P 
 
 | 
 
 CM I~ 
 
 t^ tf 
 
 or. c< 
 
 00 ci- 
 
 o oo 
 
 ft 
 
 <N C 
 
 CO C 
 
 00 CN 
 in' CO 
 
 Of p cc 
 O LO <N CC 
 
 »o lO 
 d CM 
 
 
 S 
 
 CO "" 
 
 c- co 
 
 CM CJ 
 
 co c- 
 
 -H CO 
 
 
 
 -* Tt 
 
 t~ IC 
 
 
 >— I o- 
 
 
 
 d 
 
 
 (N IC 
 
 O f 
 
 CM C 
 
 CM OO 
 
 P? 
 
 a 
 
 t^ t^ 
 
 O CO 
 
 co cc 
 
 co c: 
 
 oo -*' 
 
 — r- 
 
 1-1 
 
 1-1 
 
 1-1 "" 
 
 
 o3 -P B 
 
 
 
 
 
 
 
 - c3 H 
 
 
 <N cr 
 
 —I OC 
 
 -# u- 
 
 
 r-- in 
 
 lit 
 
 
 p C 
 
 I- cr. 
 
 
 oo cc 
 
 
 <j 
 
 f OC 
 
 CO o 
 
 io on 
 
 CO t-- 
 
 oo f 
 
 CO c^ 
 
 
 <N — 
 
 co c. 
 
 
 <:t — 
 
 
 
 
 
 
 
 « 
 
 03 
 
 tf *- 
 
 o — 
 
 ^h t^ 
 
 t-H cc 
 
 ur, 03 
 
 a 
 
 t^. V 
 
 ^H Cft 
 
 p o- 
 
 CO cc 
 
 
 
 C35 CC 
 
 Hi lO 
 
 oo r- 
 
 CO c 
 
 IN f ' 
 
 
 ^ 
 
 »C -£ 
 
 ■*> lO 
 
 co ->+ 
 
 CO CC 
 
 CO lO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 "o3 
 
 "3 
 
 "3 
 
 13 
 
 15 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 
 
 (h 
 
 
 
 
 & 
 
 03 
 
 a 
 
 cS 
 
 o3 
 
 03 
 
 
 
 01 
 
 ft 
 
 ft 
 
 ft 
 
 ft 
 
 ft 
 
 
 
 02 
 
 03 
 
 o3 
 
 03 
 
 03 
 
 03 
 
 
 
 XI 
 
 X 
 
 ,C 
 
 J3 
 
 si 
 
 
 
 
 o c 
 
 w c 
 
 w c 
 
 ° C 
 
 u c 
 
 
 
 b c 
 
 (- c 
 
 U £ 
 
 h c 
 
 "2 © 
 
 
 
 d T 
 
 03 T" 
 
 
 o3 •- 
 
 03 •« 
 
 
 
 § bi 
 
 J S w 
 
 S 'S 
 
 1 § SI 
 
 § M 
 
 
 
 C a 
 
 h- 1 *- 
 
 B fi £ 
 
 
 C cp 
 
 
 
 hh -c a 
 
 HH J) 
 
 H-O 4 
 
 H "O « 
 
 
 
 g cs 4: 
 
 S 1 ^ K 
 
 2 * c 
 
 S 1 ^ w 
 
 ° C 
 
 Ml 
 
 .2 d c 
 
 
 
 « 
 
 
 tf 
 
 
 w 
 
 
 « 
 
 rt 
 
 
 
 "3 « 
 
 
 r 
 
 i 
 
 
 
 
 
 
 6 
 
 
 
 
 
 
 
 
 
 
 hi 
 
 
 +3 
 
 
 -„ 
 
 Qj 
 
 >>,a 
 
 
 
 
 ti 
 
 >-j 
 
 l/J 
 
 •- 
 
 
 
 '4= 
 
 a 
 
 
 0) 
 
 o 
 
 .5 4l.S° 
 
 5 o *" 
 
 £^ "S 
 
 ft M-C 
 
 ill 
 
 r ^i3.2 
 >>2 s 
 
 03 +-> CC 
 
 3 ^ c3 
 £ g 0i 
 
 a> u m 
 fe"- 1 g 
 
 *- 
 
 
 
 +3 
 
 
 
 i" 
 
 03 
 
 <J1 
 
 03 
 o 
 
 C- 
 
 as I-. 
 
 M 
 
 "ft 
 
 71 
 
 < 
 
 
 s 
 
 
 s 
 
 fc 
 
 > 
 
 
 - 
 
 15 
 
 pi 
 
 o 
 
 
 -• 
 
 sj 
 
 03 
 
 q 
 
 CO 
 
 5 
 
 if 
 
 
 
 Pi 
 
 
 
 
 C3 
 
 ~ 
 
 1 
 
 a. 
 
 a 
 
 
 
 
 
 C 
 
 
 91 
 
 cl 
 
 - 
 
 CO 
 
 oj 
 
 r. 
 
 C 
 
 c 
 
 
 a-. 
 
 p 
 
 03 
 
 -: 
 
 r. 
 
 
 
 
 
 ^ 
 
 
 crt 
 
 
 o; 
 
 f* 
 
 
 ~i 
 
 a- 
 
 eS 
 
 : 
 
 
 o 
 
 _ 
 
 3 
 
 — 
 
 -C 
 
 ft 
 
 
 > 
 
 CI 
 03 
 
 £ 
 
 c 
 
 u 
 
 o 
 
 o 
 
 
 s^'*<! 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 11 
 
 contrasted with the slender-leaved chamise. Within the broad-leaved chapar- 
 ral forms are many species whose leaves are from several to many times as 
 broad as they are thick; these include both sprouting and nonsprouting 
 species. The chamise, on the other hand, includes only one widely distributed 
 species, which bears heathlike, needle-shaped leaves in fascicles. Chamise is 
 a strongly sprouting form of chaparral. This distinction of chamise as a major 
 component of chaparral has important ecological significance, for vast chap- 
 arral lands are dominated by almost pure chamise stands, and these stands 
 are mostly found on sites of particularly low productivity. 
 
 Herbaceous plants form a sparse understory of more or less suppressed, 
 mostly annual plants in undisturbed chaparral lands. Before burning of the 
 brush, this undergrowth produces little seed, whereas after burning, the her- 
 baceous forms grow robustly, fruit freely, and increase rapidly in numbers. 
 
 CLIMATE AND SOILS OF THE CHAPARRAL 
 
 Climate. — The foothills and valleys of California are characterized by rains 
 which come late in the fall, in winter, and in early spring; by mild winter 
 temperatures, and by hot, dry summers (82, 83, 111, 112). Eighty per cent 
 or more of the yearly precipitation of the valleys and foothills occurs during 
 the winter and spring seasons. Thus the climate is distinctly "Mediterranean." 
 The yearly distribution of precipitation and temperatures in chaparral areas 
 is shown in table 1, while the regions appear in figure 4. 
 
 The rapid growth period occurs in the spring months, especially March and 
 April. At this time, the soil moisture is abundant, and air temperatures are 
 far above the minimum necessary for growth. Most chaparral species also 
 flower during those months; but in some species, such as the manzanitas, 
 flowering starts as soon as growth begins, and continues until about late May, 
 when growth for the year has practically ceased. In most species, the fruiting 
 period is completed by June, a short time before the soil moisture is exhausted. 
 
 Chaparral occurs over a comparatively wide geographical range of pre- 
 cipitation and thermal belts of similar yearly patterns (20, 75). Areas of 
 chaparral, although disrupted by areas of grass, are found from the margin 
 of the fog-belt forest of the north-coast ranges to the edge of the deserts, except 
 in the interior valleys and the higher mountains. This cover commonly occupies 
 hillsides, and occurs only occasionally in the valleys, apparently because of 
 increased depth of the soil and correspondingly stronger competition with 
 grass cover. Climate, however, seems to play the major role in delimiting the 
 principal boundaries of the chaparral association. The extremes of tempera- 
 ture and precipitation at critical periods for an extended length of time, 
 rather than the average or mean of these values, appear to be the important 
 climatic factors determining distribution of the chaparral. Thus this cover is 
 found only limitedly, if at all, in the interior valleys and the deserts, where 
 extremes of temperatures, in excess of 100° F, occur for several successive 
 days at frequent intervals every summer. This is especially true where the 
 average annual precipitation is below 14 inches. On the other hand, in locali- 
 ties where temperatures seldom exceed 100°, the chaparral may survive, even 
 though the precipitation in some years drops as low as 10 inches. In such 
 locations the stand assumes an open, desertlike aspect. 
 
12 
 
 University of California — Experiment Station 
 
 At the other moisture extreme, dense chaparral stands are rarely found 
 where the average annual precipitation is in excess of 40 inches. In such heavy- 
 precipitation zones, forest growth excludes the chaparral. 
 
 The relative humidity, and the direction and velocity of the wind, as indi- 
 cated by various local studies, appear to have little correlation with the dis- 
 tribution of chaparral, except along the coast, where wind movement is more 
 
 [ JChamise predominant 
 
 L J Broadleaf chaparral predominant 
 
 mil Mixed chamise and broadleaf chaparral 
 
 Fig. 4. — Distribution of the chaparral association by ecological regions. 
 
 or less continuous and often high. Grasses and prostrate shrubs dominate the 
 wind-swept slopes facing the ocean, and trees are the predominant forms in 
 the small protected valleys and coves along the coast. That the natural bound- 
 aries of chaparral are not rigidly defined by physical factors, but are strongly 
 influenced by competition with other plant forms, is evidenced by the aggres- 
 siveness of chaparral invasions where the vigor of neighboring plant associa- 
 tions has been reduced by grazing, cutting, or fire. 
 
 Soils. — Chaparral is tolerant of widely different soil conditions {103, 115), 
 being found on a large number of soil series. Some chaparral species occupy 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 13 
 
 serpentine soils, which support few other plant species. Most of the dense 
 stands of this hard brush, however, are found on relatively poor soils 7 where 
 other plant forms fail to grow. These soils are usually shallow and are fre- 
 quently interspersed with stones or concretions. 
 
 JAN. FEB. MAR 
 
 AUG. SEPT. OCT. NOV DEC. 
 
 Fig. 5. — Comparative mean temperatures and precipitation in the 
 five chaparral regions. 
 
 SPECIAL CHARACTERISTICS OF THE CHAPARRAL REGIONS 8 
 
 On the basis of climatic variation, and differences in floristic composition, 
 the California chaparral association may conveniently be separated into five 
 ecological regions. These regions are designated in the present discussion as 
 follows: (I) North Coastal Region, (II) Central Coastal Region, (III) South 
 
 7 The term "good" or "productive" soil is here used to denote fairly deep soil, the subsoil 
 of which is moderately to well covered with a layer of fine-textured, usually dark-colored 
 top soil, and with relatively little rock outcrop or stony surface. Various luxuriant grasses 
 and other herbs occur under the less dense brush and in the openings. 
 
 The term "poor" or "unproductive" soil is used here to denote areas whose soils are 
 coarse textured or so thin, rocky, or low in organic matter that little grass or other palatable 
 herbs will come in after the brush is burned. Serpentine or "green" soils, exposed subsoils, 
 soft or hard bedrock, shale, or stones are indicative of unproductive soils. Little herbaceous 
 growth is to be found under the brush or in the open spaces. 
 
 8 Indebtedness is here expressed to the California Forest and Eange Experiment Station 
 of the United States Forest Service, and particularly to A. E. Wieslander, in charge of the 
 cover-type-map compilation, for permission to use original field maps in the construction of 
 this brush-distribution map, and for information on the vegetative characteristics of vari- 
 ous species. Additional data for plotting were secured from the Geology Department of the 
 California State Division of Mines, and from measurements and observations taken in the 
 field. 
 
 Where distribution of the brush was questionable, as in certain northern counties, the 
 present writer made a general survey to determine and verify boundary lines. Acreage was 
 computed from planimeter readings taken from a large-scale map upon which the brush 
 areas had been plotted as accurately as possible from the data available. Small islands 
 separated from the larger stands of brush, which could not be shown on the small-scale map, 
 have been disregarded in this computation. Moreover, jagged borders of the scrub cover, 
 when transferred to the smaller-scale map, were of necessity smoothed out in many instances. 
 
14 University of California — Experiment Station 
 
 Coastal Region, (IV) North Sierran Region, and (V) South Sierran Region 
 (fig. 4) . Each of these regions has in common the same dominant growth forms, 
 but there are differences in topography and, to some extent, in distinctive 
 species. Differences in climate are also evident, as shown in table 1 and figure 5. 
 Climatic averages and percentages were computed from the data of 346 
 county, state, and federal weather-reporting stations. 
 
 North Coastal Region. — This region extends northward from San Francisco 
 Bay to Trinity and Shasta counties (fig. 4, 1). Approximately the northern 
 one fourth of this region, and the brush areas in the mid western portion of 
 Mendocino and Sonoma counties, support mostly broad-leaved chaparral, 
 with only limited stands of mixed chamise-chaparral. The remainder of the 
 region is composed of chamise and of mixed chamise-chaparral, in approxi- 
 mately equal proportions. The valley side of the coast ranges is dominated by 
 chamise, whereas the coastal side of the range supports mostly mixed chamise- 
 chaparral. Whiteleaf manzanita is conspicuous on the serpentine areas of 
 Napa and Lake counties, and in the extreme northern portion of the broad- 
 leaved chaparral association. The dominant chaparral species appearing in 
 the region are listed below : 
 
 Sprouting species Nonsprouting species 
 
 California scrub oak Common manzanita 
 
 Chamise Hoary manzanita 
 
 Eastwood manzanita Stanford manzanita 
 
 Interior live oak Wedgeleaf eeanothus 
 
 Leather oak Whiteleaf manzanita 
 Western mountain-mahogany 
 
 The climate of this region varies from the comparatively cool, humid coastal 
 fog belt, to the drier region of the inner coast ranges, where the summer tem- 
 peratures frequently exceed 100° F. The average temperature values for 
 January are 39°, 44°, and 46° degrees, respectively, for weather stations 
 immediately above, within, and immediately below the chaparral zone; and 
 for July the average temperatures of the respective stations are 70°, 72°, and 
 73°. The corresponding mean annual temperatures are 53°, 57°, and '59°. 
 
 The average winter precipitation is the heaviest of the five chaparral 
 regions, being 25, 18, and 16 inches above, within, and below the chaparral 
 zone, respectively. The average annual precipitation is 46, 31, and 27 inches 
 in these same zones. The average summer precipitation is less than 1 inch in 
 any of the three zones. For seasonal divisions, see footnote of table 1. 
 
 The shallow Konokti soil series — a heavy clay loam soil of granular struc- 
 ture and with neutral to slightly acid reaction — supports the densest stands 
 of chamise and chaparral in the region. 
 
 Central Coastal Region. — This region extends from the San Francisco Bay 
 area, southward through the coast ranges to the Santa Ynez and San Rafael 
 mountain ranges, covering Santa Barbara County and part of Ventura County 
 (fig. 4, II). In most of this region chamise predominates, but chamise and 
 broad-leaved chaparral occur along the coastal side of the ranges in Monterey 
 and San Luis Obispo counties, and along the coast facing the Santa Barbara 
 Channel in Ventura and Santa Barbara counties. Much of the broad-leaved 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 15 
 
 chaparral is restricted to serpentine soils. The dominant chaparral species 
 appearing in the region are listed below : 
 
 Sprouting species Nonsprouting species 
 
 California scrub oak Bigberry manzanita 
 
 Canyon live oak Jim brush 
 
 Chamise Parry ceanothus 
 
 Chaparral whitethorn Wartleaf ceanothus 
 
 Eastwood manzanita Wedgeleaf ceanothus 
 Greenbark ceanothus 
 Interior live oak 
 
 The average temperature values for January are 46° and 49° F, respec- 
 tively, for weather stations within and immediately below the chaparral zone ; 
 and for July the average temperatures of the same stations are 70° and 65°. 
 The mean annual temperatures are 56° and 58° within and below the chap- 
 arral, respectively. 
 
 The winter precipitation of this region is light, averaging 11 inches within 
 the chaparral and in the zone below; and the average annual precipitation is 
 18 and 19 inches, respectively, for these zones. No rainfall, other than traces, 
 has been recorded during the summer in the chaparral, and an average of only 
 0.2 inch has fallen below the chaparral ; the region is practically without snow. 
 
 Chaparral is the dominant vegetation throughout, there being no forests 
 except in small protected valleys along the coast, and on some of the higher 
 peaks where the precipitation is greater. 
 
 There are fewer valleys than in the North Coastal Kegion, but some, like 
 the Salinas Valley, are more extensive. Also there is a wide variety of soils, 
 most of which have been derived from sedimentary or mixed parent materials. 
 In some localities poor drainage and alkali are serious deterrents to plant 
 growth. The extensively occurring sandy or loamy-sand soils of this region 
 are perhaps the most easily eroded of California soils. 
 
 South Coastal Region. — This region extends through the counties of Ven- 
 tura, Los Angeles, and San Bernardino, south to Lower California (fig. 4, III) . 
 The brush is confined to the coastal ranges, and to the San Bernardino Moun- 
 tains, the area to the east being largely desert. The native vegetation is pre- 
 dominantly chamise. A rather extensive area of intermingled chamise and 
 broad-leaved chaparral cover, however, occurs in the San Gabriel and San 
 Bernardino mountains, and an area of predominantly broad-leaved chaparral 
 extends south through the Cuyamaca Range, from Riverside County into San 
 Diego County. Numerous small areas of broad-leaved chaparral are also 
 found along the lower limits of the extensive chamise belts. The dominant 
 chaparral species are listed below : 
 
 Sprouting species Nonsprouting species 
 
 Canyon live oak Bigberry manzanita 
 
 Chamise Bigpod ceanothus 
 
 Chaparral whitethorn Cupleaf ceanothus 
 
 Eastwood manzanita Hairy ceanothus 
 
 Interior live oak Parry manzanita 
 
 Mission-manzanita Wartystem ceanothus 
 
 Bibbonwood Wedgeleaf ceanothus 
 Western mountain-mahogany 
 
16 University of California — Experiment Station 
 
 The average temperature values for January are 38°, 46°, and 52° F, re- 
 spectively, for weather stations immediately above, within, and immediately 
 below the chaparral zone ; and for July the average temperatures of the same 
 weather stations are 70°, 73°, and 73°. The average annual temperatures for 
 these stations are 54°, 59°, and 62°. 
 
 This unit of the chaparral association is marked by low precipitation. The 
 average winter rainfall is 19, 11, and 9 inches for weather stations above, 
 within, and below the chaparral, respectively ; and the average annual pre- 
 cipitation for the same stations is 39, 22, and 16 inches. An average of 1 inch 
 of rainfall has been recorded during the summer months above and within the 
 chaparral, whereas mere traces of rainfall have been recorded in the zone 
 below the chaparral. Rainfall appears to be the most decisive factor in limit- 
 ing the extent of the brush cover in this region, but here the most extensive 
 and diverse hard-brush areas of the state occur. The brush is apparently a 
 climax, or permanent cover, over the greater portion of these foothills. 
 
 The topography of this region is extremely irregular, extending from sea 
 level to more than 8,000 feet, a condition which gives rise to many soil types. 
 Several of the soil series have an "iron" hardpan beneath the loamy coarse 
 sand. Most of these soils are relatively infertile; hence they support only 
 sparse vegetation. 
 
 North Sierran Region. — This region extends from Butte and Tehama coun- 
 ties, southward along the Sierra Nevada foothills to Mariposa County (fig. 4, 
 IV). The chaparral occurs intermittently within open forests and is composed 
 of isolated stands, compared with the extensive areas of the coast ranges. North 
 of Bear River the brush fields are essentially composed of broad-leaved chap- 
 arral, whereas south of the river, a combination of chamise and broad-leaved 
 chaparral predominates. The dominant chaparral species of this region are 
 given below : 
 
 Sprouting species Nonsprouting species 
 
 Brewer oak Mariposa manzanita 
 
 California scrub oak Wedgeleaf ceanothus 
 
 Chamise Whiteleaf manzanita 
 
 Indian manzanita 
 Toyon 
 
 Western mountain-mahogany 
 Woollyleaf ceanothus 
 
 This is essentially a region of localized climates, modified by topography. 
 The precipitation and temperature factors for the region, and for the adjoin- 
 ing areas, are summarized as follows : The average temperatures for January 
 are 41°, 44°, and 46° F, respectively, for weather stations immediately above, 
 within, and immediately below the chaparral zone; and for July the average 
 temperatures of the same weather stations are 73°, 74°, and 79°. The mean 
 annual temperatures are 56°, 59°, and 60° for the corresponding stations. 
 
 Here the winter precipitation is 26, 17, and 11 inches, respectively, for sta- 
 tions above, within, and below the chaparral, whereas the average annual 
 precipitation is 49, 32, and 20 inches, respectively. During the summer months, 
 an average of approximately 1 inch of rainfall is received above the chaparral ; 
 none has been recorded within or below the chaparral zone. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 17 
 
 Much of the annual precipitation comes as snow, the yearly distribution 
 being somewhat similar to that of the north-coast ranges ; but here the average 
 annual precipitation in the brush areas may vary from 10 to 60 inches in 
 different years in the same localities. Also the extremes of temperature are 
 greater than in any other chaparral area of the state. 
 
 South Sierran Region. — This region forms a rather narrow strip along the 
 foothills, through Tulare County, and into central Kern County (fig. 4, V). 
 Its nearest chaparral connection toward the south is in the South Coastal 
 Region in Los Angeles County ; on the north the chaparral is separated from 
 the North Sierran Region by Madera and Fresno counties, in which there are 
 no extensive brush fields. The northern portion of the region, in Tulare 
 County, has a cover of chamise and chamise-chaparral in about equal propor- 
 tion, whereas the southern portion, in Kern County, has no chamise and is 
 made up entirely of a broad-leaved chaparral cover. The dominant chaparral 
 species appearing in this region are as follows : 
 
 Sprouting species Nonsprouting species 
 
 Brewer oak Mariposa manzanita 
 
 California scrub oak Wedgeleaf ceanothus 
 
 Chamise 
 
 Interior live oak 
 
 Western mountain-mahogany 
 
 Because of aridity and high temperatures, this region is not especially fa- 
 vorable to the growth of brush, as the following summary shows. The average 
 atmospheric temperatures for January are 40°, 45°, and 47° F, respectively, 
 for weather stations immediately above, within, and immediately below the 
 chaparral zone; and for July the temperatures are 73°, 82°, and 83° for the 
 same stations. The mean annual temperatures are 54°, 61°, and 63° for the cor- 
 responding stations. 
 
 The precipitation is limited in this region ; only 9, 10, and 6 inches of pre- 
 cipitation is received during the winter months at stations above, within, and 
 below the chaparral region, whereas the annual mean precipitation for these 
 stations is 18, 16, and 11 inches, respectively. Only traces of rainfall have been 
 recorded in these three zones during the summer months. 
 
 The average precipitation is the lowest, and the average length of the grow- 
 ing season the shortest, of these factors in the five chaparral regions. Unlike 
 the short growing season in the North Sierran Region, however, which is 
 limited primarily by frequent frosts, the growing season of the South Sierran 
 Region is severely limited by high temperature and low moisture availability. 
 
 The soils are similar in most respects to those of the North Sierran Region, 
 except that there are fewer well-developed profiles. The more shallow, rocky 
 soils are predominantly derived from acid igneous rock. There are also a few 
 scattered areas with soils of basic igneous derivation. The Vista soil series, 
 which is neutral in reaction, and is derived from acid igneous rock, is most 
 characteristic of the chaparral areas in this region. 
 
 To summarize : The data presented indicate that temperature and precipi- 
 tation are the most important factors in determining the general distribution 
 of the chaparral cover. Extended periods of cold and snow in the winter, 
 or several consecutive days of temperatures higher than 100° F recurring 
 
18 University of California — Experiment Station 
 
 throughout the summer, appear important in limiting the distribution of the 
 hard-brush association. Annual precipitation below approximately 10 inches 
 is evidently insufficient to maintain chaparral, and annual precipitation over 
 35 to 40 inches is apparently more favorable to forest growth which usually 
 replaces the chaparral in such climatic areas. 
 
 SURVEY OP VIEWPOINTS AND PRACTICES 
 CONCERNING BRUSH BURNING 
 
 A survey of past and present management practices prevailing in chaparral 
 lands influenced the planning of certain aspects of the studies reported here. 
 Although conflicting interests and viewpoints have made the practice of 
 brush burning a controversial one, outspoken expression of these points of 
 view has perhaps exaggerated the problem. Among the viewpoints frequently 
 supported by various factions is that of the favorable effect of the Indians' 
 holding the brush in suppression. Thus it is contended by some that until fires 
 were curtailed by state laws an acute brush-invasion situation did not exist. 
 A review of the burning activities of the Indians, and of the state fire-control 
 policies, together with a consideration of experiences and attitudes of stock- 
 men concerning brush burning, is presented in the following pages. 
 
 BURNING BY INDIANS IN CALIFORNIA 
 
 According to some advocates of broadcast burning, the Indians of Califor- 
 nia burned vegetation so frequently and extensively before white men settled 
 in the region that the brush was kept down, and the timber and woodland 
 kept open. Others claim that most of the fires were accidental, and not pur- 
 posely set. Since the assertions of widespread burning by the Indians have 
 not hitherto been studied closely, this phase of the present investigation was 
 undertaken in an attempt to bring together the most reliable documentary 
 evidence on the following questions : Why, where, how extensively, and under 
 what conditions did the Indians burn grassland, chaparral, and forest? 
 Answers to these questions are summarized here, according to the most reliable 
 documentary evidence available. 
 
 "Broadcast burning" has become a common term, and is used here to denote 
 burning of standing brush or other cover, with or without the adequate safe- 
 guard of fire breaks, and with or without the presence of an adequate crew of 
 ranchmen during the burning. 
 
 The term "controlled broadcast burning" is here used to imply so-called 
 "sanctioned" burning of standing brush or other vegetation through the grant- 
 ing of a permit, issued by the State Division of Forestry. The season and actual 
 time of burning, location of firebreaks, and size of the crew deemed necessary 
 to control the fire are stipulated or recommended in the regulations. 
 
 Critical review of the mass of documents published from 1542 to about 1853 
 leads to the conclusion that California Indians burned vegetation, limitedly 
 at least, to facilitate hunting, to secure native plant foods, and to clear small 
 areas of woody vegetation for the growing of tobacco. But these same docu- 
 ments indicate that the fires were seldom extensive. Kroeber (56), Barrett, 9 
 
 9 Barrett, L. A. A record of the forest and field fires in California from the days of the 
 early explorers to the creation of the forest reserves. Official records of U. S. Dept. Agr., 
 Forest Service, Region 5, San Francisco. 1935. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 19 
 
 and Barrett and Gifford (7) incline to the belief that burning by Indians 
 was somewhat general in California, but that by far the most extensive and 
 destructive fires have occurred since the coming of the white man. An indica- 
 tion that Indians probably did not burn extensively is suggested by the fact 
 that only a small proportion of the more than five hundred narratives ex- 
 amined even mentioned fires in grassland, brushland, or forest. 
 
 Some Indians, notably those of the central Sierra Nevada, are reported 
 to have cleared the underbrush with fire to facilitate travel, hunting, and 
 planting. Gordon-Cumming (28), for example, stated in 1883 that the In- 
 dians in the Yosemite Valley burned some areas so frequently as to keep 
 down the woody undergrowth. The Miwok Tribe, which inhabited the eastern 
 edge of the grassland, and whose territory extended through the brush and 
 into the lower part of the tree zone, burned grassland somewhat frequently, ac- 
 cording to Barrett and Gifford (7). These Indians probably did little to pre- 
 vent the spread of fire to the trees, and some of the fire-scarred older trees may 
 have come about from such grass burning (28) . Drucker, 10 however, concluded 
 that in Riverside, Orange, and San Diego counties the Indians did not set fire 
 to the brush or trees. Nothing was found in the literature to refute this. 
 
 A few chroniclers reported Indian burning as locally destructive. Thus 
 Costanso (21), who was on the Portola expedition to San Francisco Bay in 
 1769-1770, reported that at one point near the present Palo Alto, the progress 
 of the expedition was halted because of the absence of pasture where the 
 natives had burned. Palou (72), with the same expedition, reported several 
 burns along the coast. Of one place in Monterey County, he stated that "the 
 soil is whitish and short of pasture on account of fires set by the heathen." 
 
 In a critical review of this manuscript, however, the late Dr. C. Hart Mer- 
 riam, who devoted much of his life to the study of California Indians, stated 
 in 1935 that he did not recall having heard of the use of fire in agriculture by 
 the Indians except to clear the tiny spots used as seedbeds for tobacco. This 
 conclusion is corroborated by Kroeber (56), who also points out that the In- 
 dians of the northwestern part of California practiced some degree of con- 
 trolled burning in the preparation of seedbeds for small patches of tobacco. 
 
 The use of fires to drive rabbits, deer, and other game into the open, or into 
 ambush, is reported by a number of writers (7, 10, 22, 46, 52, 53, 69, 100). 
 Before the valleys were settled by white men, the native tribes living in the 
 grassland area burned small circular areas occasionally to collect grasshop- 
 pers and small game (5, 11, 33, 44, 53, 77, 105) . 
 
 Fire was also used, to some extent by the Indians as a means of obtaining 
 plant food. Hinds (40) and Belcher (12) of H.M.S. Sulphur noticed in 1837 
 some trees and stands of small shrubs burning along the Sacramento River. 
 They mentioned that the natives set fire to the bases of large oak trees, thereby 
 destroying some of them while attempting to get acorns. Helper (34) reported 
 that the Digger Indians, when driven by extreme hunger, collected acorns by 
 burning down oak trees the bark of which was used as a storage chamber of 
 these fruits by birds. Kroeber 11 reported some burning in oak groves for the 
 
 10 Drucker, Philip. Field notes taken in 1934-1935. Official records of Department of 
 Anthropology, University of California, Berkeley, Calif. 
 
 11 Kroeber, A. L., Professor of Anthropology, University of California, in a memorandum 
 to the author dated September 3, 1941. 
 
20 University of California — Experiment Station 
 
 purpose of clearing ground so that acorns could be collected more easily. 
 Baxley (9) observed that Indians in Yosemite Valley burned small areas to 
 facilitate gathering of their winter supply of acorns and wild sweet-potato 
 roots. Gibbs (27) reported that Indians burned grass spots along the Russian 
 River so that aniseed might be collected with greater ease. 
 
 Kroeber concluded that the Indians burned "considerably" in both open 
 country and forest for various reasons, but he expressed the opinion that "this 
 burning was not indiscriminate, but tended to be limited to certain tracts in 
 which they were interested." This worker furthermore concluded : "In gen- 
 eral, the Indians nowhere burnt the chaparral with the idea of getting rid of 
 it. A good stand of it is harder to get through after burning than before. If they 
 fired the chaparral occasionally it was with the idea of driving the game out." 
 Kroeber also affirmed that "the aggregate extent [of California lands] burned 
 over occasionally or more or less regularly must have been considerable," but 
 he expressed uncertainty as to whether burning in brushlands was extensive. 
 Various other references (9,12) indicate that Indian burning of brushlands 
 was on such a restricted scale that it could have influenced little, if at all, the 
 present composition, or the distribution, of the chaparral over the state. This 
 conclusion is supported by the fact that the major Indian population was 
 along the coastal slopes and in the valleys, away from the present distribution 
 of the chaparral lands. A compilation by Kroeber (56), giving the geographi- 
 cal ranges and population of tribes before the coming of the white man, shows 
 the distribution to be as follows : 
 
 Indian 
 Area population 
 
 Coastal 62,500 
 
 Valley 54,500 
 
 Plateau 8,500 
 
 Mountain 7,500 
 
 Total 133,000 
 
 Areas remote from the coastal and valley regions evidently were relatively 
 little subject to directed or systematic burning. It is probable, therefore, that 
 at least a fair (if not the major) proportion of the fires reported by the early 
 explorers at the higher elevations were started by lightning, as they are today. 
 
 Review of historical documents indicates that the California Indians fre- 
 quently used fire temporarily to clear various kinds of lands, including some 
 chaparral areas, for different purposes. Since the few Indians who lived in 
 the brushfields were apparently interested chiefly in local or "spot" burning 
 to facilitate hunting of deer, it seems doubtful that they burned frequently 
 or extensively enough in the chaparral association to keep much of the cover 
 open, or that such burning resulted in either expansion or contraction of the 
 brush cover as a whole. Evidently areas remote from the coastal and valley 
 regions were little subject to Indian burning. A large proportion of the fires 
 reported by early explorers, particularly in the higher elevations, were pre- 
 sumably started by lightning. Accordingly, study of Indian burning in Cali- 
 fornia is historically interesting, but of little application in the present-day 
 effort of brush suppression. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 21 
 
 HISTORY OF STATE FIRE-CONTROL POLICIES PERTAINING 
 TO BRUSHLANDS 
 
 Although burning of vegetation antedates the advent of the white man, it 
 was apparently not until after the gold rush of 1849 that burning was entered 
 into on an extensive scale. Legislative recognition of the effects of fire is 
 first found in a proclamation against burning practices issued by Governor 
 Jose Joaquin Arrillago on May 31, 1793. The statement comments on the seri- 
 ous damage done by fires started by "Christians and gentiles," and suggests 
 that offenders be punished when fires are started. 
 
 Early Legislation. — The first fire legislation in California was passed in 
 1872. 12 This law made the setting of fires on state or federal lands an offense 
 punishable by fine or imprisonment. This legislation, however, was poorly 
 enforced, and consequently was of little avail in controlling burning (80). 
 
 Effective legislation against promiscuous burning awaited a more aroused 
 public consciousness, which came shortly after the creation of the national 
 forests. On March 18, 1905, an act "to provide for the regulation of fires on, 
 and the protection and management of, public and private forest lands," was 
 passed by the legislature and approved by the governor (15). Malicious or 
 negligent setting of fires on land, other than that owned by the individual, or 
 allowing fires to escape to other lands, was defined as a misdemeanor and made 
 punishable by fine or imprisonment. Burning of brush, stubble, or other vege- 
 tation on any lands without a permit was illegal between May 15 and October 
 31. Restitution of damage from fire which escaped from private lands was 
 made possible by holding the offender liable for double the amount of damage, 
 provided such escaped fires were due to neglect ; but the offender was liable 
 only for the actual damage if the fire spread from unavoidable causes. Ad- 
 ministration of the act was provided for by creation of a Division of Forestry, 
 a State Board of Forestry, and the office of the State Forester. The function of 
 the State Forester was to carry out the policies adopted by the Board (14, 15) . 
 
 Reorganization of the administration of state resources was undertaken in 
 1927, with the creation of a Department of Natural Resources. The Division 
 of Forestry was made a part of this organization under the jurisdiction of a 
 State Board of Forestry and a State Forester. At this time fire legislation was 
 changed but little ; the definitions of offenses were retained as in the original 
 law, but the punishment was made more severe. Liability for damage was set 
 as the costs of the actual damage and control of the fire. Burning on any forest 
 land, grassland, or brush area without permit was prohibited between April 
 15 and December 1 (15). 
 
 Despite the strict effort at prohibition of fire, and the efficient control sys- 
 tem of the state and federal forestry services, fire has continued to be a major 
 problem in the state. The fire problem has been further complicated by the 
 belief of many landowners that strict fire control is impracticable and unde- 
 sirable (13, 70). This view apparently applies particularly to the chaparral 
 areas of northern California (91) . 
 
 First Investigation on Controlled Burning. — The desire for more practical 
 knowledge on the question of burning was crystallized in 1930, when, on 
 
 12 For source see footnote 9, p. 18. 
 
22 University of California — Experiment Station 
 
 demand of the stockmen, the Conservation Committee of the North Coast 
 Council, California State Chamber of Commerce, requested the State Board 
 of Forestry to delegate two members of the Board to investigate "clearing of 
 brush" by burning in Lake, Mendocino, and Sonoma counties. Sheepmen in 
 those areas desired that the State Board of Forestry adopt a policy favorable 
 to livestock operations because they felt that it provided a means of main- 
 taining their ranges in a somewhat more productive state. 
 
 The conclusions of the survey were that, in general, burning appeared to be 
 detrimental, but that on limited areas burning did show some promise of 
 temporarily increasing the range carrying capacity. General burning of the 
 lands in question was considered inadvisable because of their watershed and 
 recreational values. As most of the chamise-covered areas responded poorly 
 after a fire, it was concluded that the adoption of a burning program would 
 probably be of little value as a general measure. 13 
 
 The committee on burning recommended that the Board of Forestry take 
 no formal action on the question, but that the State Forester instruct inspec- 
 tors in Lake, Mendocino, and Sonoma counties to issue burning permits on 
 areas approved by the local county farm advisor, or by some other state agri- 
 cultural agency. In this connection, a state forest ranger gave his entire time 
 to cooperation with the citizens of Mendocino County ; but since the results 
 of this undertaking were reported to be unsatisfactory, this plan was also 
 discontinued. 
 
 State-Controlled Brush-Burning Policy. — Further progress on the question 
 of burning of brushlands in public and private ownership was made on March 
 26, 1932. On this date the State Board of Forestry adopted a resolution to 
 segregate the lands under its jurisdiction into units of strict fire protection 
 as compared with those of intermediate and of limited importance in fire 
 suppression. Three vegetation zones were recognized with respect to impor- 
 tance in fire protection. Zone 1 consists of the more valuable timber and water- 
 shed lands, and also includes state parks and monuments, as well as other such 
 areas subject to serious economic loss by fire. The lands of Zone 2 are regarded 
 as "buffer" or boundary strips between the grasslands of the valleys, and the 
 forests of the higher elevations. They embrace much of the brushy cover and 
 the rougher privately owned woodland-grass areas, hence are less important 
 in the matter of fire protection. The lands of Zone 3 include the grassy valleys 
 and agricultural lands, and are chiefly of local interest. 
 
 Reports on the brush-burning cooperative policy following the 1938 field 
 season appeared to be somewhat favorable with respect to decrease in incen- 
 diarism. 14 From the upper Sacramento Valley region, and the north coast 
 area, came reports of some decrease in malicious burning. In 1939, on the 
 other hand, incendiary fires covered an unusually large acreage ; but the num- 
 ber of incendiary fires was again somewhat lower in 1940. The State Forester 
 expressed the belief that some success in reducing incendiary fires is being 
 achieved through the cooperation of the California Wool Growers' and the 
 
 13 Report by H. S. Gilman and Swift Berry on range clearing under state burning permits 
 on private lands in north coast counties outside of national forests to the State Board of 
 Forestry. November 29, 1930. 
 
 14 Report by C. G. Strickland, Deputy State Forester, to the State Forester on agricul- 
 tural and range improvement brush-burning policy. November 16, 1938. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 23 
 
 California Cattlemen's associations. 15 It is realized, however, that the data on 
 incendiarism are still meager and that before concrete recommendations can 
 be made, more accurate knowledge of the values, relative gains, and losses to 
 the present and future vegetation and soil must be obtained. Preliminary 
 information on some of these points might be obtained by contact with repre- 
 sentative stockmen. 
 
 EXPERIENCES AND ATTITUDES OF STOCKMEN CONCERNING 
 BRUSH BURNING 
 
 In the course of the present study, it seemed desirable to secure a cross 
 section of the observations and convictions of successful livestock men oper- 
 ating in or near the brushland areas of northern California counties. It was 
 hoped that such a survey would not only provide a picture of current senti- 
 ment on brush burning, but might also point to some of the more practical 
 phases of the subject especially in need of study. Accordingly, a questionnaire 
 was prepared in 1934, which formed the basis for a series of interviews with 
 stockmen in the same year. In Lake, Mendocino, Humboldt, Shasta, Tehama, 
 and Colusa counties, 85 operators were consulted. To ensure that the views 
 obtained should be those of stockmen with long tenure, broad experience, and 
 respected judgment, the farm advisor of each of the above-named counties 
 furnished a list of owners who he especially felt should be consulted. The 
 men from these lists, most of whom were large operators, filled out copies of 
 the questionnaire while being personally interviewed and were, in addition, 
 questioned in detail on related points. 18 Most of the field contacts were made 
 during the summer and autumn of 1934. The following are the chief points 
 embraced in the questionnaire : Location of ranch, and acreage of chaparral 
 and other lands ; kinds and number of stock owned ; number of stock grazed 
 before and after burning; condition of the animals; frequency of burning; 
 quality and palatability of the forage ; extent of erosion ; damage to improve- 
 ments; cost of burning; injury from snagging; and losses from poisonous 
 plants. 
 
 From 11 to 24 men were interviewed in each county except Colusa, where 
 only 3 were questioned. A total of 638,729 acres was included in the survey, 
 of which 96,000 acres were brushlands, much of which had been burned at 
 one time or another. Excluding Colusa County, the stockmen who were con- 
 sulted controlled an average of 10 per cent of the total brushlands in their 
 respective counties, 12 per cent of the sheep, and 26 per cent of the cattle. 
 The stability and success of operation among these men are indicated to some 
 extent by the fact that more than half of them have owned their ranches for 
 twenty -five to fifty years before the interviews. 
 
 There was little agreement in response to questions on the quality of forage 
 on burns. Thirty-four out of 85 of the men interviewed said that the quality 
 of forage on burns was "good," but 25 said that burning made no difference 
 in forage quality. Smaller numbers stated that the forage on burns was "fair" 
 or "poor." Thus opinion seemed to be sharply divided between two extreme 
 
 15 In a letter to the author from M. B. Pratt, State Forester, dated August 11, 1941. 
 
 16 Most of the interviews were conducted by J. O. Bridges, a graduate student of range 
 management in the Department of Forestry, University of California. 
 
24 University of California — Experiment Station 
 
 positions. The great variation in the quality of the soil and in the kinds of 
 brush doubtless accounts for the wide variation reported in the amount and 
 quality of the forage. 
 
 More general agreement was reflected in the answers to the question on the 
 condition of the animals grazed. Fifty-two men out of 85 agreed that the 
 condition of animals grazed on burns was "good," 25 men said that it was 
 "fair," whereas 6 stated that the condition was "poor." Several men pointed 
 out that the period of satisfactory forage use on most brush areas is only in 
 the spring and that these lands are commonly utilized only at that time. 
 
 Considerable hesitancy seemed to be shown by the stockmen interviewed 
 when asked to estimate changes in the original grazing capacity of their lands 
 irrespective of burning. Of the 85 men consulted, only 58 reported, and 28 of 
 these would make no grazing-capacity estimates. Those men who would hazard 
 an estimate were almost evenly divided among the opinions that the grazing 
 capacity of their ranches had decreased from the original, that it had in- 
 creased somewhat, or that it had remained constant. Some of the men who 
 reported a decline in forage attributed it to encroachment of brush. It was 
 evident that the ranchers as a group were not accustomed to considering the 
 carrying capacity of their pasture land in quantitative terms. 
 
 Similar uncertainty was noted in the responses to the slightly different 
 question as to how much the desirable feed was increased or decreased by 
 burning brushlands. Again, the answers were almost equally divided among 
 those who believed that there was no change in the amount of desirable feed, 
 that there was a slight increase, or that there was an appreciable increase in 
 desirable forage. Only 1 man out of the 76 answering believed that the increase 
 amounted to as much as 75 to 100 per cent for a considerable period of time 
 after burning, while several men believed that there was a definite decrease. 
 There was a tendency in Shasta and Tehama counties, however, for stockmen 
 to value the forage of burned areas appreciably higher than did the men of 
 other counties. 
 
 Answers to questions on the frequency of chaparral burning in the past 
 showed that by far the most prevalent practice in the counties concerned is 
 that of broadcast burning every eight to fifteen years. Few men had made a 
 practice of burning patches each year, but a few reported broadcast burning 
 at fairly regular intervals of five to ten years. Again Shasta and Tehama 
 counties were distinguished from the other counties in that none of the Shasta 
 and Tehama men reported planned or systematic burning programs. 
 
 While an appreciable number of stockmen preferred to make no statement 
 as to whether the brush was killed and the invading grass made permanent, 
 among those men who did express an opinion, the conviction seemed over- 
 whelming that the brush is not killed by fire. This majority agreement was 
 common to all the six counties surveyed. 
 
 Although the majority of stockmen reported that they had burned their 
 brushlands every eight to fifteen years, most of them seemed convinced that 
 more frequent burning would be desirable from the standpoint of increased 
 palatability and yield of forage. About one third of the answers favored burn- 
 ing every two to three years if it were possible, while another third favored 
 burning at intervals of four to five years. Smaller numbers believed that burn- 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 25 
 
 nig should take place at intervals of eight to fifteen years. A relatively small 
 group of men had been convinced by experience that burning on their land 
 was never desirable. Most of the stockmen believed that there would be enough 
 fuel in the brush for another burning by the fifth or sixth year. Indeed, a 
 few believed that fire could be run through the brushfield every year. The 
 apparently general conviction that burning increases the palatability and 
 yield of forage on brushlands is especially interesting in the light of the hesi- 
 tation and uncertainty of the stockmen in estimating benefits on their own 
 ranches. 
 
 Few of the stockmen considered erosion to be an important consequence of 
 brush fires. Nearly three fourths of the men believed that soil erosion is insig- 
 nificant in their region, both before and after burning. A few had observed 
 small increases in soil erosion the first year after burning, while a small num- 
 ber reported a considerable increase in soil loss during the first year after a 
 fire. Only 2 men of the 80 questioned on this point expressed themselves as 
 being aware of serious erosion. Part of this general failure to observe appre- 
 ciable erosion after brush fires seems to have resulted from lack of training 
 in detecting soil movement, and lack of understanding in evaluating its im- 
 portance. On numerous ranches showing advanced sheet and gully erosion, 
 the stockmen averred that the loss of surface soil was not serious. 
 
 From one third to one half of the stockmen in different counties reported 
 accidental damage to property as a result of brush burning. Grassland, fence, 
 hay and grain fields, and farm buildings were reported burned. Less than 
 half of the damaging fires were admittedly caused by stockmen. Motorists 
 and hunters were blamed for most of the damage. 
 
 There was considerable variation in the reports on damage to livestock on 
 chaparral burns. Of the 79 men consulted on damage to livestock, 49 made 
 no report on injury to lambs by snagging. Twenty-seven men reported that no 
 snagging occurred, whereas 3 men, all from Tehama County, reported that 
 some lambs were snagged. Ten men reported some wool pulled out by the dead, 
 burned stems, whereas 13 men reported no greater loss of wool than before 
 burning. Since 44 men made no report on the loss of wool, it seems likely that 
 they, too, failed to observe any such loss. Twenty-two men reported losses of 
 cattle and sheep from poisonous plants (65). It may be significant that 16 of 
 these were from Shasta County. Larkspur was blamed for more losses than 
 any other plant genus, but death camas was also blamed for some losses, espe- 
 cially sheep (66). Eleven of the 16 men reporting livestock deaths from poi- 
 sonous plants in Shasta County were unable to identify the plant or plants 
 causing the losses. Thirteen men reported losses of cattle, while only 6 reported 
 losses of sheep from poisonous plants. 
 
 Questions on the extent of firebreaks needed in the chaparral-burning oper- 
 ation, their cost, and the cost of burning, revealed that 36 men out of 80 con- 
 sulted do not intentionally burn their brushlands. Part of these men refrain 
 because they have become convinced that exclusion of fire is the best policy. 
 Others do not burn because brushland comprises only a small part of their 
 pastures. Still others are dissuaded by danger to their own or neighbors* 
 croplands and pasture areas, buildings, and fences, or by the responsibilities 
 imposed by the state laws. Most of those who practice burning reported that 
 
26 University of California — Experiment Station 
 
 only a backfire is necessary for control, but a relatively large minority reported 
 that more extensive control measures are necessary. It is interesting to note 
 that in Lake County no one believed that control measures more elaborate 
 than backfires were necessary, whereas in Tehama County 8 out of the 10 
 men consulted believed that greater precautions were necessary. 
 
 Estimates of the cost of firebreaks and of burning were given with great 
 uncertainty and hesitation. Only 6 men reported definite costs of burning. 
 These ranged from 5 to 10 cents per acre in Lake County to 5 dollars per acre 
 in brush and timber in Humboldt County. Most stockmen believed that proper 
 control of more or less isolated areas can be achieved with a small labor cost, 
 whereas in other instances the cost of burning is high. 
 
 The survey as a whole showed that 50 out of the 85 men questioned favored 
 the burning of brushfields, but emphasized the importance of protecting 
 grasslands and cultivated areas. Twenty more men recommended the burning 
 of all natural vegetation types not especially suitable for timber growth. Only 
 14 men were against burning of all kinds of natural vegetation. In view of 
 the uncertainty shown by stockmen in their answers to questions concerning 
 the effects of fire on the quality of forage, and the carrying capacity of their 
 lands, the sentiment in favor of burning was apparently based more on com- 
 mon belief than upon firsthand experience. The sentiment favoring burning 
 was especially strong in Shasta and Tehama counties. Eighteen out of the 20 
 men who favored burning of all types of woody vegetation lived in Shasta and 
 Tehama counties, whereas no one from these counties was against some form 
 of burning. The survey strongly revealed the need of critical study of plant 
 succession and related aspects of chaparral burns. 
 
 PLANT-SUCCESSION STUDIES ON CHAPARRAL LANDS 
 
 REVIEW OF RESULTS BY EARLIER WORKERS 
 
 In southern California and on various sites of severe conditions elsewhere 
 in the state, the chaparral association, especially when dominated by chamise, 
 is generally regarded as forming a climax cover, or the final stable vegetation 
 (18,75).^ ' 
 
 The climax cover is not always the most desirable for grazing. Restriction 
 of grazing on some ranges of the Wasatch Mountains of Utah, for example, 
 to favor complete reestablishment of the wheatgrass climax vegetation, re- 
 sulted in economic loss. In that connection Sampson (88) pointed out that 
 in certain covers in Utah it was most desirable to favor the next highest suc- 
 cessional stage — the needlegrass and yellow brush — while at the same time 
 maintaining a fair amount of wheatgrass. 
 
 Both the nature and the duration of changes in vegetative cover resulting 
 from burning depend on many factors, notably climate, topography, soil, 
 nature of the original vegetation, and land-use practices. Topography, 17 
 through the effect of slope and exposure, may directly affect succession subse- 
 quent to a fire by causing variation in soil temperature and soil moisture ; and 
 it may indirectly affect succession through its influence on erosion. Larsen 
 (58, 59), for example, found that steep south slopes present a serious situa- 
 
 17 Climatic factors of the chaparral association are discussed on p. 11. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 27 
 
 tion on burns in Idaho, where erosion is so pronounced in many places as to 
 prohibit establishment of all but the hardiest pioneering plants. Topography 
 also largely regulates the intensity and the extent of fires, hence determines 
 when and where islands of the original vegetation may be left for regenera- 
 tion (49) . Show and Kotok (96) concluded that topography is one of the most 
 important factors determining local distribution of brush in California. Cer- 
 tain topographic features favor hot fires, which may partly destroy the estab- 
 lished vegetation and its accumulated litter, and permit invasion by brush 
 (98). 
 
 The severity of the burn may affect succession by its influences on the exist- 
 ing stand, on the accumulated seed, and on the soil itself. Isaac (48) believes 
 that the ground cover present in original Douglas-fir forests, and the extent 
 to which it is killed out by fires, partly determine the extent to which invading 
 colonies of weed and brush species appear in the succession that follows. Some 
 of the original invading species persist, while others die out (86, 87, 88). In 
 the Pacific Northwest, exposure to direct sunlight resulted in raising tempera- 
 tures in the surface soil so high that seedling plants were killed (41, 48). 
 
 Many shrubs show remarkable capacity to sprout after a fire ; hence rapid 
 revegetation of burned areas may generally be expected after burning such 
 vegetation. Plummer (75), working with chaparral in California, noted rapid 
 recapture of the lands by chamise, and some species of chaparral, after burn- 
 ing. Cooper (20), on the other hand, concluded that whereas a single or inter- 
 mittent burning of California chaparral resulted in a crop of sprouts more 
 dense than before, yearly burning, though seldom possible because of insuffi- 
 cient fuel, tended to destroy the brush, or prevent its further invasion. This 
 worker also contended that extensive areas now dominated by grass, or by 
 scattered stands of drought-enduring trees, were formerly controlled by 
 chaparral, or had been kept in a grass cover by repeated burning, and would 
 eventually support brush if fire were eliminated. Cooper (20) further con- 
 cluded that along its border of mesophytic climax, California chaparral has 
 transgressed its normal climatic limits through invasion into the forest, 
 whereas along its bordering dry zone it has been pushed back by grass and 
 drought-enduring plants in the north, and by coastal sagebrush in the south. 
 Fire is claimed to be the causative agent in each case. Isaac (49) agrees that 
 most sprouting brush species are retarded by one fire, and some may be killed 
 out by successive fires. His evidence indicates, however, that successive fires 
 in some plant covers prolong the presence of weeds and brush. Clements (18) 
 concluded that repeated fires of the past are indicated by extensive occurrence 
 of sprouting chaparral. 
 
 Ingram (47), working on Douglas-fir cutover lands, noted that various 
 shrubs soon reclaimed the soil unless they were reburned at three- to five-year 
 intervals. This worker pointed out that the use of fire to convert forest land 
 to grazing should be carefully controlled because of the many failures re- 
 corded. Although frequent burning is necessary to suppress the brush, this 
 practice often results in heavy invasions of bracken fern, a species of no 
 grazing value. 
 
 It has been generally assumed that revegetation of burns formerly occupied 
 by a nonsprouting cover is accomplished by means of seed dispersed from 
 
28 University of California — Experiment Station 
 
 bordering areas. Little, however, has been done to determine the rate and 
 extent of brush invasions in this manner. Studies of forested lands indicate 
 that much of the seedling growth of burned areas is due to seed already on 
 the forest floor at the time of the fire (38, 41, 42, 58, 59, 67). The possibirrty 
 that an analagous situation occurs in grassland and brushfields has led to 
 investigation of the relative resistance of various seeds to heat. Thus Wright 
 (117), studying germination of several plant species common in the Sierra 
 Nevada, found that the seeds of certain chaparral forms endured much higher 
 temperatures than did those of trees or grass. As pointed out elsewhere this 
 may be one of the primary factors influencing the aggressive invasions of 
 burns by chaparral vegetation in California. 
 
 The fact that soil conditions may influence succession after burning was 
 demonstrated by Stallard (98), who found that repeated fires progressively 
 destroy much of the soil organic matter. This results in retrogression of the 
 plant succession after each fire, the area finally reaching denudation — 
 essentially comparable to primary successional conditions. 
 
 Where excessive grazing has resulted in reducing the natural grass cover 
 to a point favorable to invasion by relatively unpalatable nonspr outing brush, 
 burning may aid in the restoration of grass. Morris (67), for example, found 
 that in the Great Basin region burning or grubbing of sagebrush increased 
 some species of grasses and broad-leaved herbs. 
 
 The most outstanding fact consistently appearing in the literature on fire 
 and plant succession is that no single criterion or formula may safely be used 
 to predict the outcome. An unusually dry or wet season preceding or follow- 
 ing burning may unpredictably affect the plant population of burned areas. 
 The seed of some inconspicuous plant which has accumulated in the soil for 
 many years may germinate in large numbers under the suddenly changed 
 conditions brought about by burning, and thus result in the conspicuous pres- 
 ence of that particular species after the fire. Only as a result of careful study 
 in a particular area, taking into account all possible biotic, climatic, and topo- 
 graphic factors, can rational prediction of plant succession be made. 
 
 ORGANIZATION OF THE STUDY 
 
 In initiating the brush-burning study in northern California, two major 
 problems seemed to merit careful attention. One of these was to note the effect 
 of burning on the plant cover ; the other was to study the effects of burning 
 on the fertility of the soil. Accordingly, the principal tasks of the present 
 study were to record the changes in vegetation produced naturally on burned 
 areas in the years immediately following a fire, to determine the extent to 
 which plant succession subsequent to burning may be modified by various 
 management practices, and to determine the effects of these changes on the 
 soil. 
 
 Although plant succession on burned brushlands is modified by many local 
 influences, the results presented in the section on changes in plant cover are 
 believed to show characteristic trends, and to provide underlying principles 
 which may be applied in the management of burned chaparral lands. 
 
 As previously shown, ranchers of northern California do not generally 
 consider the effects of brush fires on the soil to be important, although there 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 29 
 
 is a feeling by many that the ashes fertilize the soil. A few of the ranchers have 
 also become aware of losses of soil fertility through erosion. Since it is gen- 
 erally conceded that soil losses are not replaceable within any reasonable time, 
 it is important that such destruction be recognized in its early stages. Accord- 
 ingly, studies were instituted to ascertain the effects of burning on certain 
 chemical soil properties. The findings reported should be helpful in gaining 
 a better understanding of the effect of fire on the soil, and in counteracting 
 the menace of resultant losses. 
 
 Inquiry was also made into some of the changes in environment which might 
 influence succession. Thus the temperatures which prevail at various depths 
 of soil during a fire invited study because of their probable direct effects on 
 seed germination and survival of sprouts. The sudden, widespread appear- 
 ance of certain plants on burns, coincident with the disappearance of unre- 
 lated forms, suggests that the seeds of some plants were more tolerant to 
 extremes of heat than others. 
 
 Efforts of ranchers to influence plant succession after a fire induced the 
 writer to include various trials in reseeding. To determine the effectiveness 
 of such practices, numerous experimental reseeding trials were undertaken, 
 using many cultivated plants and some native species, under various brush- 
 land conditions. 
 
 HERBACEOUS PLANTS COMMON TO CHAPARRAL AREAS 
 
 It seemed desirable in the present study to compare the species and abun- 
 dance of herbaceous plants commonly found on recent burns, with the forms 
 which normally occur on mature unburned stands of chaparral. 
 
 Extensive surveys revealed that the species listed below make up the major 
 herbaceous cover both on burned and on unburned chaparral areas. 
 
 Common Herbaceous Plants of Brush Areas of Northern California 18 
 
 Broad-leaved herbs 
 
 Bishop lotus (m) 
 Blue gilia (p) 
 Bolander bedstraw 
 Calif ornia everlasting (m) 
 California figwort (m) 
 California filago (m) 
 California goldenrod 
 Cantua spurge 
 Chaparral cotton weed (m) 
 Chaparral pentstemon (m) 
 Coast larkspur (w) 
 Common mullein 
 Common peppergrass 
 Common soap-plant (p) 
 Common yellow mustard 
 Coyote tobacco 
 Crown brodiaea 
 Cut-leaved thelypodium (m) 
 Dotseed plantain 
 
 Dove lupine (m) 
 
 Dwarf Bridges grassnut 
 
 English plantain 
 
 Field suncup (p) 
 
 Fineleaf lotus (ra) 
 
 Fireweed (m) 
 
 Foothill death camas 
 
 Fremont death camas (m) 
 
 Fremont globem allow (w) 
 
 Gamble weed 
 
 Gold fern 
 
 Gold-wire (m) 
 
 Grassnut (m) 
 
 Greenbrier 
 
 Hill lotus (p) 
 
 Horned snapdragon (p) 
 
 Horseweed (m) 
 
 Indian pink 
 
 Indian tobacco 
 
 18 Plants marked (ra) are moderately persistent, and often fairly abundant, on recently 
 burned chaparral lands; those marked (p) are persistent and often very abundant on such 
 lands. 
 
30 
 
 University of California — Experiment Station 
 
 Broad-leaved herbs 
 
 Leafy gilia (m) 
 Little-bill loco 
 Longleaf filago 
 Meadow death camas 
 Menzies larkspur (m) 
 Milk aster 
 Moth mullein 
 Mountain nievitas (m) 
 Nada stickleaf (m) 
 Napa star thistle (2?) 
 Narrowleaf soap-plant (m) 
 Northern rockcress (m) 
 Nuttall bedstraw (m) 
 Prickly lettuce (m) 
 Purple nightshade (w) 
 Eattlesnake weed (ra) 
 Eed larkspur (m) 
 Eedmaids 
 Eed ribbons (w) 
 Eedstem filaree 
 Eush lotus 
 
 Scouler St. Johnswort 
 Sheep sorrel (w) 
 Silverleaf lupine 
 
 Bent-head fescue 
 Blue wild-rye (m) 
 California melic 
 DoAvny chess (ra) 
 Fotxail fescue (w) 
 Junegrass 
 
 Little quaking grass (m) 
 Malpais bluegrass 
 Mouse barley (ra) 
 Nitgrass (w) 
 
 Grasses 
 
 (cont.) 
 
 Slender buckwheat (p) 
 Slender madia (m) 
 Small-flowered lotus (m) 
 Small meadow death camas 
 Spanish-clover (w) 
 Summer cottonweed (ra) 
 Sword fern (w) 
 Tobacco mimulus (w) 
 Tomcat clover (ra) 
 Torrey nievitas (w) 
 Turkey-mullein (ra) 
 Valley tassels (m) 
 Western lupine (m) 
 Western thistle 
 Whispering bells (p) 
 White everlasting (p) 
 White-flowered navarretia (ra) 
 White forget-me-not (ra) 
 White ha wkweed (m) 
 Whitestem filaree 
 Wild carrot (p) 
 Wool-mat 
 Woolly yarrow (p) 
 Yellow star thistle (w) 
 
 Pacific fescue 
 Perennial ryegra ss 
 Purple needlegrass 
 Eat-tail fescue ( m ) 
 Eed brome (w) 
 Eipgut 
 
 Silver hairgrass (m) 
 Six-weeks fescue (m) 
 Slender oat (m) 
 Spanish brome (m) 
 Wild oat 
 
 Only 16 per cent of the species included in the foregoing list of plants which 
 become conspicuous after fires tend to persist in comparatively large numbers 
 five years or more after burning, although these plants may also gradually 
 decrease in abundance as the burn becomes older. Most of the species decrease 
 sharply in population by the third year, so that by the fourth or fifth year 
 after burning they are little or no more abundant than on unburned brush- 
 lands. Few of the common grasses occur in abundance after about the third 
 year, except on nonsprouting-chaparral areas. Habit sketches of the more 
 abundant common broad-leaved herbs appear in figures 6 to 10, and of the 
 commoner grasses in figure 11. 
 
 RESULTS OF PLANT-SUCCESSION STUDY 
 
 The most conspicuous change in the vegetation may be expected where a 
 dense woody cover is largely or completely burned. Since the post-burning 
 successional stages are mainly composed of annual or short-lived perennial 
 herbs, the cover may fluctuate conspicuously in species and density from year 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 31 
 
 to year, with development towards the original relatively stable brush vege- 
 tation in the form of sprouts and chaparral seedlings. Such development is 
 usually rapid. 
 
 RECOVERY OF BRUSH AFTER BURNING 
 
 The different kinds of chaparral vary greatly in the time required to reoc- 
 cupy the soil after burning. If the stand is composed predominantly of the 
 much more restricted nonsprouting forms, the seedling brush species will not 
 completely recapture the soil for several years after burning; and even then 
 many openings will often occur where the reproduction fails to gain a foot- 
 hold. In the much more extensive sprouting stands of chamise, and in mixed, 
 
 Tig. 6 — A, Deerweed, a half -shrub. B, Bush monkey-flower, and C, gold-wire — both 
 herbaceous. All are common on burned areas. 
 
 sprouting chaparral, however, recovery after burning is generally rapid and 
 complete, and may extend well beyond the pre-fire boundaries of the burn. 
 The rate of growth and recovery of sprouting brush stands has a significant 
 bearing on subsequent treatment of the lands. In order to obtain quantitative 
 data relative to the annual growth rate of such extensive brush covers, all of 
 the brush plants were cut at the surface of the ground, and weighed, from a 
 series of plots varying in size from one milacre to 20 feet square. 19 The measure- 
 ments were recorded in September, well after completion of the season's 
 growth, in 1937, 1938, and 1939, on burns from one to eight years old. The 
 cut brush of each plot was tied in bundles of a size convenient for weighing 
 with a steelyard scale beam. The dry weights of the samples representing each 
 age class were averaged, and the results are presented graphically in figure 12. 
 
 19 The growth-recovery study was made in Mendocino County, to represent coastal con- 
 ditions, and in Shasta County to typify the interior brush region, on burns of known history 
 and age. Ten to 13 milacre plots and 5 to 7 random plots of 20 feet square were cut and 
 weighed for each age class in each county. The dry weight of samples was computed from 
 oven-dry composite samples consisting of proportional parts of the coarse stems and the 
 leafy branches. 
 
32 
 
 University of California — Experiment Station 
 
 Fig. 7. — Five broad-leaved herbs common to chaparral and chamise areas: A, woolly- 
 3 r arrow; B, leafy gilia; C, chaparral penstemon; D, Torrey nievitas; D-2, enlargement of 
 tip of stem of Torrey nievitas ; and E, blue gilia. 
 
 Some of these sample materials were analyzed for their contents of certain 
 mineral constituents. 
 
 Table 2 shows that recovery of chamise and of sprouting chaparral stands 
 is rapid for several years after a burn. The average dry weights of chamise per 
 acre, for example, one and five years after burning, were 1,560 pounds and 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 33 
 
 Fig. 8. — One of the most conspicuous broad-leaved herbs on burned chaparral areas is 
 common soap-plant (A). Others are: B, common madia; C, nada stickleaf; D, field suncup; 
 and E, Fremont death camas. 
 
 8,920 pounds, respectively. A similar trend is indicated in broad-leaved chap- 
 arral forms (figs. 13 and 14). In the sixth year after burning the rate of 
 recovery of both chamise and chaparral covers was slower; and the annual 
 increment declined appreciably in the eighth year. When total dry weight 
 
34 
 
 University op California — Experiment Station 
 
 Fig. 9. — A, Whispering bells, one of the most abundant and characteristic broad-leaved 
 herbs on chaparral burns. Other abundant forms are : B, hill lotus, with detailed flowering 
 branch shown at B-2 ; C, longleaf filago; D, rattlesnake weed, with detailed fruiting branch 
 shown at D-2 ; and E, Napa star thistle. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 35 
 
 Fig. 10. — Some broad-leaved herbs common to chaparral and chamise areas: A, California 
 figwort; B, tobacco mimulus; C, purple nightshade; D, horned snapdragon; and E, red 
 ribbons. 
 
 per acre is plotted against time in years (fig. 12), an almost straight line of 
 increment is produced during the early years, indicating a rapid and nearly 
 constant annual rate of growth for a period of four to six years after burning. 
 The more flattened trend of the curves after the sixth year indicates an appre- 
 ciable decline in growth rate. 
 
36 
 
 University of California- — Experiment Station 
 
 Fig. 11. — Grasses common to chaparral and chamise areas: A, foxtail fescue; 
 B, red brome ; C, silver hairgrass ; D, nitgrass. 
 
 From the standpoint of brushland management, the chief significance of 
 the growth-rate trend is that it shows the approximate period of maximum 
 growth in sprouting brush cover; this indicates when replacement of the 
 herbaceous understory vegetation is most rapid. The poundage of fuel present 
 at stated intervals after burning is also shown in the table. The amount of fuel 
 
Bul. 685 J Plant Succession on Burned Chaparral Lands 
 
 37 
 
 may be of some value as an index to the probable reaction of heat on the 
 seed and the soil. 
 
 Rapid growth rate of brush has proved to be distinctly a deterrent to maxi- 
 mum production of herbaceous vegetation after the second year because of 
 shading, and the composition for soil moisture by sprouts and seedlings of 
 the chaparral vegetation (figs. 13 and 14). The curve shown in figure 12 indi- 
 cates rapid growth of sprouting chaparral covers, with its consequent reduc- 
 tion in herbaceous vegetation, until the sixth year. Even so, experience has 
 shown that in order to get a fairly clean burn, firing should not be done more 
 frequently than at about seven- or eight-year intervals. Burning at shorter 
 
 ■T3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 - „-~-'~ 
 
 
 
 
 
 
 
 ^^ 
 
 "" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 /l 
 /l 
 
 
 
 
 
 
 
 
 
 / 1 
 
 / 
 / 
 
 
 
 
 
 
 
 
 / s 
 
 
 
 
 
 
 
 
 
 / / 
 
 / s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 / 
 / 
 
 
 
 Chan 
 
 -lisp 
 
 
 
 
 ' *•«*""' 
 
 
 
 
 Chap 
 
 arral 
 
 
 
 
 
 
 
 
 
 
 
 3 4 5 6 
 
 YEARS AFTER BURNING 
 
 Fig. 12. — Annual rate of recovery of sprouting brush after burning, 
 expressed in pounds of ovendry weight per acre of top growth. 
 
 intervals often results in only partial burning of the chaparral, or in leaving 
 numerous patches of brush unburned. 
 
 The ash liberated by burning has some effect on plant succession, as is shown 
 later. Table 2 also gives the approximate amount of potassium, calcium, and 
 phosphorus released by burning. A series of plot measurements showed that 
 about 30 per cent, by weight, of shrubby growth is merely charred rather than 
 reduced to ashes by the ordinary brush fire, parts of most of the coarser stems 
 remaining. This 30 per cent of remaining unburned stems was taken into 
 account in computing the probable total yield of ash in a brush fire. The 
 percentage values used in computing these yields of liberated salts were 
 determined by analysis of several oven-dried samples. The following data 
 embrace both chamise and mixed growth of chaparral : 
 
 Mineral Chamise Chaparral 
 
 Potassium, per cent 0.640 
 
 Calcium, per cent. 0.709 
 
 Phosphorus, per cent . , 0.062 
 
 Total, per cent 1.411 
 
 0.730 
 0.920 
 0.073 
 
 1.723 
 
38 
 
 University of California — Experiment Station 
 
 The weight of the different elements subject to liberation as ash on burned 
 chaparral lands (table 2) is seen to be approximately the same as that on 
 chamise areas. Also there is little difference in the total amounts of salts in 
 the two types of covers. 
 
 As pointed out elsewhere (p. 83), evidently only a small fraction of the 
 liberated ash is ordinarily available for plant growth on a newly burned area. 
 On hillsides, particularly, a large part of the newly liberated ash is blown 
 into drainage channels where it is washed away. Of that which remains on 
 
 TABLE 2 
 
 Dry Weights of Chamise and Chaparral Covers of Different Age Classes, 
 
 and Calculated Minerals Liberated by Burning 
 
 
 Plot 
 area 
 
 Chamise 
 
 Chaparral 
 
 County 
 
 Dry 
 weight 
 per acre, 
 pounds 
 
 Mineral constituents 
 
 available for liberation 
 
 by burning, 
 
 pounds per acre 
 
 Dry 
 weight 
 per acre, 
 pounds 
 
 Mineral constituents 
 
 available for liberation 
 
 by burning, 
 
 pounds per acre 
 
 
 Potas- 
 sium 
 
 Cal- 
 cium 
 
 Phos- 
 phorus 
 
 Potas- 
 sium 
 
 Cal- 
 cium 
 
 Phos- 
 phorus 
 
 One-year-old burns 
 
 Mendocino. 
 
 Lake 
 
 Lake 
 
 Shasta 
 
 Shasta 
 
 Shasta 
 
 Shasta 
 
 Average... 
 
 1,630 
 1,820 
 1,170 
 2,250 
 1,260 
 1,030 
 1,780 
 
 1,660 
 
 7.3 
 8.2 
 5 2 
 10.1 
 5.6 
 4.6 
 8.0 
 
 7.0 
 
 8.1 
 9.0 
 5.8 
 11.1 
 6 2 
 5 1 
 
 7.7 
 
 0.7 
 8 
 5 
 1.0 
 5 
 0.4 
 0.8 
 
 0.7 
 
 1,220 
 1,460 
 1,040 
 1.010 
 1,000 
 1,130 
 1,020 
 
 1,180 
 
 5.7 
 
 7.9 
 9.4 
 6.7 
 5.0 
 6.5 
 5.6 
 
 0.6 
 0.7 
 0.5 
 5 
 0.5 
 0.6 
 0.5 
 
 
 
 Two 
 
 -year-old burns 
 
 
 
 
 
 
 
 1 
 
 2 
 3 
 4 
 5 
 6 
 
 2,680 
 3,270 
 1,830 
 2,000 
 3,010 
 3,790 
 
 IS, 760 
 
 12.0 
 14.6 
 8.2 
 8.9 
 13.5 
 17.0 
 
 12.4 
 
 13.4 
 16.2 
 
 9.1 
 9.9 
 14.9 
 18.8 
 
 13.7 
 
 1.2 
 1.4 
 0.8 
 0.9 
 1.3 
 1.6 
 
 1.2 
 
 1,030 
 1,640 
 2,590 
 1,430 
 2,360 
 1,870 
 
 1,820 
 
 5.2 
 
 8.4 
 13.2 
 
 7.3 
 12 1 
 
 9.6 
 
 11.8 
 
 5.1 
 8.1 
 16.6 
 7.1 
 15.2 
 12.1 
 
 10.7 
 
 0.5 
 
 
 1.6 
 
 
 1.3 
 
 
 0.7 
 
 
 1.2 
 
 
 1.0 
 
 
 1.0 
 
 
 
 Three-year-old burns 
 
 
 1 
 
 2 
 3 
 4 
 5 
 6 
 7 
 8 
 
 3,960 
 5,070 
 3,750 
 4,180 
 3,930 
 3,430 
 4,030 
 4,870 
 
 4,150 
 
 17.7 
 22.2 
 16.8 
 18.7 
 17.6 
 15.3 
 18.0 
 21.8 
 
 18.6 
 
 19.6 
 25.1 
 18.6 
 20.7 
 19 5 
 17.0 
 20.0 
 24.2 
 
 20.6 
 
 1.7 
 2.2 
 1.6 
 1.8 
 1.7 
 1.5 
 1,7 
 2.1 
 
 1.8 
 
 5,260 
 3,980 
 3,070 
 4,980 
 2,730 
 4,620 
 3,000 
 3,260 
 
 8,860 
 
 26.9 
 20.3 
 15.7 
 25.5 
 13.9 
 23.6 
 15.3 
 16.7 
 
 19.7 
 
 33.9 
 25.6 
 19.4 
 32.1 
 17.6 
 29.8 
 19.3 
 21.0 
 
 24.8 
 
 2.7 
 
 
 3.2 
 
 
 1.6 
 
 
 2.5 
 
 
 1.4 
 
 
 2.4 
 
 
 1.5 
 
 
 1.7 
 
 
 t.l 
 
 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 39 
 
 
 
 TABLE 2 — (Continued) 
 
 
 
 
 
 
 Plot 
 area 
 
 Chamise 
 
 Chaparral 
 
 County 
 
 Dry 
 
 weight 
 per acre, 
 pounds 
 
 Mineral constituents 
 
 available for liberation 
 
 by burning, 
 
 pounds per acre 
 
 Dry 
 
 weight 
 per acre, 
 pounds 
 
 Mineral constituents 
 
 available for liberation 
 
 by burning, 
 
 pounds per acre 
 
 
 Potas- 
 sium 
 
 Cal- Phos- 
 cium phorus 
 
 Potas- 
 sium 
 
 Cal- 
 cium 
 
 Phos- 
 phorus 
 
 Four-year-old burns 
 
 
 1 
 
 2 
 3 
 4 
 5 
 6 
 
 6,190 
 5,970 
 7,400 
 6,030 
 7,980 
 7,000 
 
 6,600 
 
 27.7 
 26.7 
 33 2 
 27 
 31.3 
 31.4 
 
 29.6 
 
 30 7 
 29 6 
 36.7 
 29 9 
 34 6 
 34 .7 
 
 32.7 
 
 2.7 
 
 2 6 
 
 3 2 
 2.6 
 3 
 3 
 
 2 9 
 
 4,860 
 5,670 
 7,000 
 5,320 
 4,970 
 5,840 
 
 5,610 
 
 24.9 
 29.0 
 
 35 8 
 27.2 
 25.4 
 29 8 
 
 28.7 
 
 31.3 
 36.5 
 45.1 
 34.3 
 32.0 
 37.6 
 
 36.1 
 
 2.5 
 
 
 2.9 
 
 
 3.6 
 
 
 2 7 
 
 
 2.5 
 
 
 3.0 
 
 
 2.9 
 
 
 
 Five-year-old burns 
 
 
 1 
 
 2 
 3 
 4 
 5 
 
 6 
 7 
 
 7,990 
 9,200 
 8.370 
 9.270 
 7,930 
 9,250 
 10,420 
 
 8.920 
 
 35 8 
 41 2 
 37.5 
 41.5 
 35.5 
 41.5 
 49.5 
 
 40.4 
 
 39 6 
 45.6 
 41 5 
 46 
 39.3 
 45.9 
 51.8 
 
 3 5 
 
 4 
 
 3 6 
 4.0 
 3.4 
 4.0 
 
 4 5 
 
 S.9 
 
 7.970 
 8.730 
 8,840 
 9,260 
 9,870 
 8,690 
 9,770 
 
 9,020 
 
 40.7 
 
 44.6 
 
 45 1 
 47.3 
 50.4 
 44 4 
 49.9 
 
 46 
 
 39 5 
 56 2 
 43.8 
 59.6 
 49 
 56 
 48.4 
 
 50.3 
 
 4 1 
 
 
 4 5 
 
 
 4.5 
 
 
 4.7 
 
 
 5 
 
 
 4 4 
 
 
 5 
 
 
 4.6 
 
 
 
 Six-year-old burns 
 
 
 1 
 
 
 
 3 
 4 
 5 
 
 6 
 
 7 
 8 
 
 10,240 
 9,250 
 11,330 
 11,220 
 10,020 
 9,940 
 8,760 
 9,920 
 
 10,080 
 
 45.9 
 41.4 
 49.9 
 49.8 
 49.3 
 44.5 
 39.2 
 44.5 
 
 454 
 
 50.8 
 45.8 
 56.2 
 55 6 
 49.7 
 49 3 
 43.4 
 
 49 2 
 
 50 
 
 4.4 
 4 
 4.9 
 4.9 
 4 3 " 
 4.3 
 3.8 
 4.3 
 
 44 
 
 8,930 
 10,860 
 
 8,930 
 10,440 
 
 9,380 
 
 8,480 
 10,940 
 
 9,740 
 
 9.710 
 
 45.6 
 55.5 
 46.0 
 53.3 
 47.9 
 43 3 
 55.9 
 49.8 
 
 49.6 
 
 57.5 
 
 69.9 
 57.5 
 67.2 
 60.4 
 54.6 
 70.4 
 62.8 
 
 62.6 
 
 4.6 
 
 
 5.5 
 
 
 4.6 
 
 
 5.3 
 
 
 4.8 
 
 
 4.3 
 
 
 5.6 
 
 
 5.0 
 
 
 5.0 
 
 
 
 Seven-year-old burns 
 
 
 1 
 
 2 
 3 
 4 
 5 
 
 6 
 
 9,340 
 10,830 
 10,760 
 11,930 
 11,390 
 10,760 
 
 10,840 
 
 41.9 
 48.5 
 48.2 
 53.5 
 51.0 
 48.2 
 
 48.5 
 
 46.3 
 53.7 
 53.4 
 59.2 
 56.4 
 53.4 
 
 53.7 
 
 4.1 
 4.7 
 4.7 
 5.2 
 4.9 
 4.7 
 
 4.7 
 
 9,130 
 11,270 
 8,920 
 9,860 
 10,930 
 10,570 
 
 10,115 
 
 46.6 
 57.6 
 45.6 
 50.4 
 55.9 
 53.9 
 
 57.7 
 
 58 .8 
 72 6 
 57.5 
 48.8 
 70.4 
 52.4 
 
 60.0 
 
 4.7 
 
 
 5.7 
 
 
 4.6 
 
 
 5.0 
 
 
 5.6 
 
 
 5.4 
 
 
 5.2 
 
 
 
40 
 
 University of California — Experiment Station 
 
 
 
 TABLE 2— (C 
 
 oncludt 
 
 id) 
 
 
 
 
 
 
 Plot 
 area 
 
 Chamise 
 
 Chaparral 
 
 County 
 
 Dry 
 weight 
 per acre, 
 pounds 
 
 Mineral constituents 
 
 available for liberation 
 
 by burning, 
 
 pounds per acre 
 
 Dry 
 
 weight 
 per acre, 
 pounds 
 
 Mineral constituents 
 
 available for liberation 
 
 by burning, 
 
 pounds per acre 
 
 
 Potas- 
 sium 
 
 Cal- 
 cium 
 
 Phos- 
 phorus 
 
 Potas- 
 sium 
 
 Cal- 
 cium 
 
 Phos- 
 phorus 
 
 Eight-year-old burns 
 
 Mendocino. 
 Mendocino. 
 
 Lake 
 
 Lake 
 
 Shasta 
 
 Shasta 
 
 Averaje. 
 
 10,370 
 11,060 
 11,900 
 12,800 
 10,090 
 12,060 
 
 11,380 
 
 46.5 
 49.5 
 53 3 
 57.3 
 45.2 
 54.0 
 
 50 9 
 
 51.4 
 54.8 
 59.0 
 63.5 
 50 1 
 59.8 
 
 56.4 
 
 4.5 
 
 4.8 
 5.2 
 5 5 
 4.4 
 5.2 
 
 12,030 
 10,741 
 11,690 
 
 9,020 
 10,070 
 
 9,950 
 
 10,580 
 
 61.5 
 54.9 
 
 59.7 
 46.1 
 51.4 
 50.8 
 
 54.0 
 
 59.7 
 69.2 
 58.0 
 
 44.8 
 64.8 
 64 
 
 6.1 
 5 5 
 6.0 
 4.6 
 5 1 
 5.1 
 
 5.4 
 
 
 Unburned unusually dense 
 
 luxuriant old stands 
 
 
 
 
 
 1 
 2 
 3 
 4 
 
 25,660 
 37,940 
 29,230 
 19,480 
 
 28,070 
 
 114.9 
 170 
 130.9 
 
 87.2 
 
 125.8 
 
 127.2 
 188.0 
 144.8 
 96.5 
 
 139.1 
 
 11.1 
 16.5 
 12.7 
 8.4 
 
 12.2 
 
 30, 150 
 36,010 
 21,950 
 22,540 
 
 27660 
 
 153 3 
 184 
 112 1 
 115.1 
 
 141.1 
 
 194.1 
 231.9 
 141 3 
 225 .4 
 
 198.2 
 
 15 4 
 
 
 18 4 
 
 
 11.2 
 
 
 11 5 
 
 
 14.1 
 
 
 
 the burned area, a fair proportion of the potassium fraction, being readily 
 soluble, would presumably be leached away by the winter rains, but part of it 
 would certainly be utilized by the vegetation. Moreover, fixation of the calcium 
 and phosphorus, and the insolubility of the phosphorus in soils having alkaline 
 reaction, would appear to leave relatively little of these two elements avail- 
 able for immediate plant nutrition. 
 
 INVASION OF HERBACEOUS PLANTS 
 
 The kinds and combinations of the more dominant herbaceous plants that 
 occupy an area after burning, and the number of years that they continue to 
 occur abundantly on the area, determine the success or failure of a burning 
 venture for grazing. In order to procure reliable data on plant succession in 
 diversified burned brushlands, a series of experimental plots were established, 
 on burned and unburned areas, in both chamise and chaparral cover, occupy- 
 ing a variety of soil types. 30 
 
 Chamise Cover. — The first study was conducted on chamise lands in Mendo- 
 cino County. The records were averaged for 26 milacre plots, located on slopes 
 
 20 The techniques employed included line and belt transects, which traversed both burned 
 and adjoining unburned areas ; and rectangular plots varying in size from a milacre to 
 several acres. Refined grazing reconnaissance methods were employed on the larger plots. 
 To assure use of identical areas for subsequent measurement over a long period of time, 
 the plots were marked by driving numbered iron stakes at diagonal corners of their bound- 
 aries, after which the plots were located on a map by means of compass and chain. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 41 
 
 and flats. The plots were located mostly on the Hugo, Los Osos, and Aiken soil 
 series, including a good cross section of variation in soil depth. The results 
 appear to be typical of chamise areas, as revealed in more extensive work later 
 in interior counties. The area was heavily burned in September, 1927. Species 
 which constituted less than 1 per cent of the cover during the years of observa- 
 
 v V1H 
 
 ** ^Hl^Jy ' f 1jJ< 
 
 
 
 | ' .;. \* -" 
 
 ; '...•'"•:. "; ., . p- .^ ffj 
 
 \ x %m 
 
 .:.,: '."■;■. ",■•.:.■ ; ' : '"'''. ; ' ."; ';. '.. ,■■■•■■>.-■■ .■ ■• " ; ■ - ■■■ .■-.•'-.■ , : 
 
 
 ' -/' \*rf- /,»#>"■ 
 
 irfe* • ,V ; »l'w X %<Wk*'' * ^ 
 
 4" iC» 
 
 
 * v»* ** ^a 
 
 ' x "'"■' ",:■ " i 0-\ 
 
 
 
 . ; 
 
 
 !||g|| 
 
 
 •*' • f 
 
 '".•■p^- T ' v //"*%"/ 
 
 
 ^ ■:{:. W% 
 
 Fig. 13. — A, One-year-old burn of greenleaf manzanita which is rapidly re- 
 occupying the soil. B, Two-year-old burn of interior live oak and Lemmon 
 ceanothus, with areas of grass occupying the openings. 
 
 tion, in this and in the other brush areas studied, are not listed in the data here 
 presented. The results are summarized in table 3. 
 
 Of the 16 species of broad-leaved herbs recorded in the 26 plots after burn- 
 ing, all but 2 were found at least limitedly on the sample plot areas before 
 burning, as shown by the data for 1927. Much the heaviest stands of broad- 
 leaved herbs occurred during the second and third years after the fire. A sharp 
 decline in the abundance of these plants occurred thereafter until in 1932, the 
 fifth year after burning. Several of the herbaceous species approached in 
 sparseness those of the unburned area of 1927, although about three times more 
 
42 University of California — Experiment Station 
 
 herbaceous vegetation was present than before the fire. Much of this cover, 
 however, was unpalatable to stock. In contrast, the maximum density two 
 years after the fire was seventeen times that of the pre-fire density, and the 
 grazing capacity was appreciably increased in the better sites. Conspicuous 
 downward trends in the amount and density of herbaceous vegetation oc- 
 curred in the third, fourth, and fifth years after burning. 
 
 Because of their abundance, robust growth, and showy inflorescence, woolly- 
 yarrow, whispering bells, chaparral cottonweed, and two species of death 
 camas were perhaps the most conspicuous herbaceous species during the first 
 two years after burning. 
 
 Fig. 14. — Greenleaf manzanita cover. The upper part of the slope (left) was 
 not burned for fifteen years, and is virtually devoid of understory vegetation. 
 The lower slope (right) shows abundant, vigorous sprouts, and some herba- 
 ceous plants. The estimated dry weight of the unburned brush was 11,000 
 pounds; of brush on the burn, 1,300 pounds. 
 
 The grasses, all but one species of which were annuals, were the most abun- 
 dant numerically of the herbaceous forms, both before and after burning. 
 They were most abundant during the first two years following the fire, declin- 
 ing thereafter somewhat gradually and reaching the lowest point in 1932, the 
 fifth year after burning. Silver hairgrass, nitgrass, and foxtail fescue were 
 by far the most abundant. These species yield lightly, mature early in the 
 season, and decline rapidly in nutrition. 
 
 Perhaps the most significant effect of the fire was the heavy invasion of 
 brush seedlings. No young plants of the dominant chamise, or of greenleaf 
 manzanita, were found on any of the plots before burning, whereas chamise, 
 in particular, was conspicuously abundant on all plots the first year after 
 burning, and many were well established by the fifth year after the fire. These 
 seedlings, and a strong growth of brush sprouts, evidently are primary factors 
 in the decline of the herbaceous vegetation, as they occupy much of the top- 
 growth space. A review of the plant record of these plots in 1934, the seventh 
 year after the fire, revealed that a large proportion of the brush seedlings 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 43 
 
 TABLE 3 
 
 Average Number of Understory Plants per Milacre Plot in Chamise Cover before 
 Burning in 1927, and at Yearly Intervals Thereafter; Mendocino County* 
 
 Species 
 
 Before 
 burning 
 
 After burning 
 
 First 
 year 
 
 Second 
 year 
 
 Third 
 year 
 
 Fourth 
 year 
 
 Broad-leaved herbs 
 
 Fifth 
 year 
 
 Bolander bedstraw 
 
 Bush beardtongue 
 
 Chaparral cotton weed. . . . 
 Common peppergrass .... 
 
 Common soap-plant 
 
 Fineleaf lotus 
 
 Foothill death camas 
 
 Fremont death camas .... 
 
 Menzies larkspur 
 
 Napa star thistle 
 
 Rattlesnake weed 
 
 Redstem filaree 
 
 Turkey-mullein 
 
 Whispering bells 
 
 White-flowered navarretia 
 Woolly-yarrow 
 
 Total 
 
 0.2 
 4 
 0.2 
 1.1 
 5 
 11 
 
 0.1 
 2 
 0.3 
 
 0.2 
 1.3 
 1 
 3 
 1.4 
 
 7.4 
 
 2.6 
 3.5 
 13.8 
 
 7.8 
 2.7 
 9.7 
 1.4 
 5.9 
 3.8 
 6.3 
 4.2 
 5.2 
 8.4 
 17.2 
 6.1 
 6.8 
 
 105 4 
 
 5.0 
 3.6 
 
 10.8 
 3.8 
 3.9 
 
 11.7 
 2.4 
 7.2 
 5.4 
 
 11.3 
 2.6 
 7.1 
 
 11.9 
 
 21.3 
 7.4 
 7.9 
 
 123.3 
 
 3.4 
 2.5 
 
 8.1 
 3.1 
 2.2 
 4 9 
 1.5 
 3 4 
 4.8 
 
 3.7 
 7.6 
 2.6 
 6.2 
 
 1 
 
 1 
 
 5 
 
 2 
 
 2 
 
 3. 
 
 1 
 
 1. 
 
 2. 
 
 4. 
 
 0.0 
 
 1.7 
 
 2.4 
 
 2 4 
 
 3.1 
 
 2.4 
 
 38.5 
 
 2.2 
 1.2 
 0.0 
 
 2.2 
 
 2.0 
 0.3 
 2.1 
 2.6 
 4.1 
 0.0 
 2.1 
 1.9 
 1.3 
 1.7 
 2.1 
 
 25.8 
 
 Grasses 
 
 
 6.6 
 9.2 
 7 
 3.6 
 
 6.2 
 
 2.8 
 
 46 3 
 64.2 
 14 
 16 3 
 83.7 
 12.1 
 
 57.5 
 78.4 
 1.0 
 24.8 
 97.3 
 15 4 
 
 34.8 
 29.2 
 
 1.0 
 19.6 
 51.6 
 
 8.4 
 
 22.4 
 
 21.6 
 
 1.0 
 
 11.3 
 42.8 
 10.1 
 
 19.4 
 
 Nitgrass 
 
 Purple needlegrass 
 
 Red brome 
 
 12.3 
 1.0 
 
 8.5 
 38.6 
 
 Wild oat 
 
 6.1 
 
 
 
 Total 
 
 29.1 
 
 224.0 
 
 274.4 
 
 144.6 
 
 109.2 
 
 85.9 
 
 
 
 Brush seedlings 
 
 
 
 1.0 
 0.0 
 
 0.3 
 12 
 04 
 
 47.6 
 5 4 
 7.2 
 4.2 
 4.2 
 3.1 
 
 28.0 
 3.8 
 4 
 4.7 
 5.4 
 3.1 
 
 14.3 
 3.4 
 2.6 
 3.2 
 3.5 
 2.2 
 
 12.2 
 2.4 
 2.6 
 2.1 
 3.3 
 2.2 
 
 12.2 
 
 Deerweed 
 
 Greenleaf manzanita 
 
 2.4 
 2.6 
 2.1 
 
 
 2.4 
 
 Wedgeleaf ceanothus 
 
 2.1 
 
 Total 
 
 1.46 
 
 71.7 
 
 49.0 
 
 29.2 
 
 24.8 
 
 23.8 
 
 
 
 Average of 26 plots. 
 
 present in 1932 had survived, and that the sprouts had developed into robust 
 plants. 
 
 Several plots established on chamise burns of various ages in Lake, Shasta, 
 and Tehama counties, from 1926 to 1937, gave results similar to those pre- 
 
44 
 
 University of California — Experiment Station 
 
 sented in table 3 with respect to the identity and permanence of the invading 
 herbaceous species, and particularly the establishment of chamise sprouts and 
 seedlings. These observations lend weight to the conclusion that the succes- 
 sional behavior outlined is fairly typical of burned chamise areas throughout 
 
 Fig. 15. — Heavy invasion of verba santa on the two-year-old chamise burn. 
 The chamise sprouts and seedlings eventually suppress this competing growth. 
 El Dorado County. 
 
 Fig. 16. — Chamise plot, burned in 1934, upon which succession and soil 
 erosion were recorded: A, the plot after burning and installation of run-off 
 measurement equipment; B, the same plot as it appeared in 1936. The her- 
 baceous vegetation, consisting mostly of weeds, was being crowded out by the 
 chamise sprouts and seedlings. Shasta County. 
 
 northern California. In some localities one group of herbs would be much in 
 evidence, in other localities another group of herbs would dominate ; but all 
 were much the same in aspect and successional reaction. In all instances, 
 sprouts of chamise, manzanita, poison oak, and frequently yerba santa, ap- 
 peared early in the season of the first year after burning (fig. 15) . The sprouts 
 and brush seedlings continued to develop in subsequent years until little other 
 vegetation remained (fig. 16) ; but in the interim the carrying capacity was 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 45 
 
 TABLE 4 
 
 Average Number of Understory Plants per Milacre Plot in Sprouting Manzanita and 
 
 Ceanothus Cover, before Burning in 1927, and at Yearly Intervals 
 
 Thereafter; Lake and Mendocino Counties* 
 
 Species 
 
 Before 
 burning 
 
 After burning 
 
 First 
 year 
 
 Second 
 year 
 
 Third 
 year 
 
 Fourth 
 year 
 
 Fifth 
 year 
 
 Broad-leaved herbs 
 
 
 0.3 
 0.3 
 0.4 
 0.3 
 0.6 
 0.8 
 0.4 
 0.2 
 1.4 
 2 3 
 1.0 
 2.1 
 4 
 1.2 
 0.3 
 15 
 1 
 1 
 1.3 
 
 1.4 
 3.9 
 3.8 
 3.9 
 2.7 
 2.8 
 3.4 
 4.8 
 6.4 
 5.8 
 4.6 
 5.9 
 3.1 
 9.3 
 2.3 
 12.3 
 8.3 
 3.7 
 5.7 
 
 2.6 
 5.1 
 4.1 
 3.7 
 4.6 
 2.3 
 2.6 
 4 1 
 5.8 
 4.3 
 2.3 
 4.6 
 2.5 
 4.2 
 2.1 
 13.7 
 5.7 
 6.4 
 7.9 
 
 1.8 
 2 3 
 2.9 
 1.4 
 1.8 
 1.7 
 3.1 
 3.2 
 2.9 
 2.6 
 2.4 
 5.8 
 2 3 
 3.1 
 1.6 
 4.8 
 2.8 
 4 1 
 4.0 
 
 54.6 
 
 1.2 
 1.4 
 0.4 
 0.9 
 0.9 
 1.6 
 2.2 
 0.8 
 3.1 
 2.7 
 1.6 
 3.1 
 1.6 
 1.4 
 1.2 
 1.9 
 1.6 
 3.4 
 2.8 
 
 33.8 
 
 1.4 
 
 
 6 
 
 Fineleaf lotus 
 
 2 
 
 
 4 
 
 Gamble weed 
 
 1.2 
 
 
 7 
 
 Grassnut 
 
 2 4 
 
 Nada stickleaf 
 
 1 
 
 N uttall bedstraw 
 
 2.2 
 
 
 3.1 
 
 Red ribbons 
 
 0.5 
 
 Redstem filaree 
 
 2.6 
 
 Small meadow death camas 
 
 0.9 
 1.6 
 
 
 0.4 
 
 
 1.2 
 
 Western thistle 
 
 1.2 
 
 
 1.9 
 
 
 2.6 
 
 
 
 Total 
 
 15.0 
 
 94.1 
 
 88.7 
 
 25.2 
 
 
 
 Grasses 
 
 Brush seedlings 
 
 Foxtail fescue 
 
 8.2 
 0.7 
 
 5.3 
 2.8 
 4.6 
 3 
 1.6 
 1.4 
 
 29.8 
 
 2.6 
 
 31.7 
 
 11.6 
 
 9.1 
 4.1 
 
 18.7 
 5.1 
 
 47 3 
 
 3.8 
 42.1 
 
 7.8 
 14.3 
 
 6.5 
 12.6 
 
 6.7 
 
 27.7 
 1.4 
 19.2 
 10 4 
 11.4 
 3.8 
 15.2 
 2.4 
 
 18.3 
 1.2 
 
 21.3 
 5.6 
 8.3 
 5.1 
 7.9 
 1.7 
 
 14.9 
 
 
 9 
 
 Nitgrass 
 
 11.6 
 
 Red brome 
 
 4.8 
 
 
 9.2 
 
 
 2.8 
 
 
 5.4 
 
 Wild oat 
 
 2.1 
 
 
 
 Total 
 
 24.9 
 
 112.7 
 
 141.1 
 
 91.5 
 
 69.4 
 
 51.7 
 
 
 
 
 0.0 
 0.0 
 0.0 
 
 0.0 
 0.0 
 0.0 
 
 4 4 
 
 3.7 
 9.7 
 16.4 
 1.7 
 4.3 
 
 3.8 
 2.1 
 7.6 
 
 12.2 
 2 8 
 2.3 
 
 3.3 
 1.6 
 
 7.2 
 
 10.5 
 
 1.2 
 
 14 
 
 2.3 
 1.4 
 7.2 
 10 1 
 9 
 1.4 
 
 23 3 
 
 2 3 
 
 
 1.4 
 
 
 7.2 
 
 
 10.1 
 
 
 0.9 
 
 
 14 
 
 
 
 Total 
 
 
 
 40.2 
 
 31.8 
 
 25.2 
 
 23 3 
 
 
 
 *Average of 19 plots. 
 
46 University of California — Experiment Station 
 
 increased on areas of productive soils, whereas on thin soils little or no im- 
 provement was obtained. Occasionally areas several acres in extent were over- 
 run with deerweed, an unpalatable species that sometimes grew so densely and 
 vigorously for two or three years after the fire as to appear to replace the brush 
 sprouts. By the fourth or fifth year after burning, however, the deerweed 
 usually became heavily infested with dodder, which in turn yielded its space 
 to the taller, more persistent hard brush. 
 
 Sprouting Manzanita-Ceanothus Cover. — The sprouting, broad-leaved 
 chaparral cover, composed mainly of species of manzanita and ceanothus, 
 with a subdominance of dwarf interior live oak, chaparral, coffeeberry, and 
 yerba santa, was most intensively studied in Lake and Mendocino counties. 
 These studies were later amplified by obtaining data on various plots in several 
 interior counties. The Los Osos, Konokti, Hugo, and Aiken soil series were 
 represented. A summary of the plant population of 19 milacre plots, mapped 
 yearly from 1928 to 1932, inclusive, is presented in table 4. 
 
 Although this association of sprouting chaparral contained a fairly large 
 number of herbaceous forms, namely, 19 species of broad-leaved plants and 8 
 species of grasses, the latter vegetation occupied only slightly more than 1 per 
 cent of the ground cover before burning. Maximum density of herbaceous 
 plants occurred the first and second years after burning, when the broad- 
 leaved forms and the grasses were approximately nine and five times more 
 abundant, respectively, than before burning. In the third, fourth, and fifth 
 years after burning, the decline in density of these understory plants was 
 sharp. By the fifth year, although the understory herbaceous plants were more 
 numerous than before burning, the stand was so scattered as to furnish little 
 forage. By this time the chaparral seedlings and sprouts had largely recap- 
 tured the areas. 
 
 That fire stimulated germination of the seed of brush species is impressively 
 shown by the entire absence of brush seedlings before burning, as contrasted 
 with the large number present after burning. The fact that only slightly less 
 than two thirds of the number of chaparral seedlings recorded the first year 
 after burning survived in the fifth year, despite the exceedingly heavy growth 
 of sprouts, is indicative of the aggressive and persistent character of this 
 cover. Moreover, in the fifth year after burning, the new plants fruited, illus- 
 trating the strong reproductive capacity and persistence of this vegetation. 
 
 Nonspr outing Manzanita-C eanothus Cover. — Areas covered with the much 
 less extensive nonspr outing chaparral, composed also mostly of various species 
 of manzanita and ceanothus, react differently to burning than do lands occu- 
 pied by sprouting forms of these genera. The fact that nonsprouting brush is 
 killed by a single heavy burn appears to account for the relatively rich her- 
 baceous flora that appears and usually persists well after burning on such 
 areas (fig. 17). 
 
 The initial plots of nonsprouting chaparral, established in Mendocino 
 County in 1927, were located on various slopes and flats of the Aiken, Mari- 
 posa, and Los Osos soil series. In general these soils were comparable in qual- 
 ity with more productive lands of the chamise and sprouting chaparral areas 
 whose successional results are summarized in tables 3 and 4. The results of 
 the plant study of the nonsprouting cover are summarized in table 5. 
 
Bul. G8J 
 
 Plant Succession on Burned Chaparral Lands 
 
 47 
 
 The table brings out two points which are in distinct contrast to those 
 recorded in the chamise and sprouting chaparral vegetation. First, a much 
 larger number of herbaceous species typically inhabited the nonsprouting 
 cover, both before and after burning, possibly because the brush tends to be 
 
 Fig. 17. — A, Nonsprouting manzanita burned two years previously, show- 
 ing heavy growth of ungrazed annual grasses and a few young manzanita seed- 
 lings. B, Nonsprouting manzanita burned seven years previously, showing 
 growth of six-year-old fruiting brush plants. Eeburning of this area would 
 greatly improve the grazing capacity by removing the dead brush and destroy- 
 ing much of the younger brush. Shasta County. 
 
 somewhat more open ; and second, there was an increase in the density of the 
 herbaceous species continuing even into the fifth year after burning. Al- 
 though early-maturing annuals, such as foxtail fescue, rat-tail fescue, nitgrass, 
 and silver hairgrass, were relatively abundant both before and after burning, 
 such species as soft chess and slender oat reached maximum abundance in the 
 fifth year after the fire. The increase in grass was partly offset, however, by the 
 sharp decline in most of the broad-leaved herbaceous species, and the rapid 
 
TABLE 5 
 
 Average Number of Understory Plants per Milacre Plot in Nonsprouting 
 
 Manzanita and Ceanothtts Cover before Burning in 1927, and at Yearly 
 
 Intervals Thereafter ; Mendocino County* 
 
 Species 
 
 Before 
 burning 
 
 After burning 
 
 First 
 year 
 
 Second 
 year 
 
 Third 
 
 year 
 
 Fourth 
 year 
 
 Fifth 
 year 
 
 Broad-leaved herbs 
 
 
 0.3 
 0.2 
 0.2 
 0.2 
 1.2 
 0.5 
 0.6 
 4 
 1.1 
 0.0 
 4 
 2 
 
 0.4 
 0.3 
 0.4 
 0.3 
 1.3 
 0.0 
 1.2 
 0.6 
 0.2 
 1.3 
 0.2 
 5 
 0.3 
 1 
 4 
 0.7 
 
 0.8 
 0.6 
 1.6 
 2.4 
 4.6 
 2.5 
 2.7 
 4.9 
 2.9 
 2.4 
 3.7 
 1.3 
 2.1 
 2.3 
 2.2 
 1.2 
 2.8 
 2.7 
 1.8 
 4.1 
 5.9 
 8.1 
 3.6 
 4.1 
 7.2 
 1.3 
 0.6 
 3.8 
 4.6 
 
 4 
 
 0.4 
 0.3 
 1.6 
 
 10.9 
 1.4 
 3.5 
 2.3 
 1.7 
 1.0 
 2.1 
 0.8 
 1.6 
 0.2 
 3.7 
 2.1 
 1.3 
 2.1 
 0.6 
 2.8 
 6.6 
 4.1 
 1.2 
 
 17.3 
 4.9 
 0.5 
 4 
 4.6 
 
 16.4 
 
 1.2 
 0.6 
 0.1 
 0.8 
 3.8 
 0.8 
 2.1 
 4 
 0.4 
 0.1 
 0.7 
 1.1 
 2.0 
 0.0 
 1.3 
 1.3 
 2.2 
 1.6 
 1.2 
 1.6 
 3.8 
 5.9 
 0.6 
 8.3 
 3 2 
 0.2 
 0.6 
 5.1 
 5.7 
 
 0.4 
 0.6 
 
 0.3 
 
 4.2 
 0.6 
 2.1 
 
 
 0.2 
 
 0.4 
 1.4 
 0.0 
 0.6 
 0.4 
 1.5 
 2.8 
 0.9 
 0.3 
 2.7 
 2.3 
 1.4 
 4.2 
 0.9 
 0.0 
 0.3 
 4.3 
 4.9 
 
 37.7 
 
 0.3 
 
 
 0.3 
 
 
 
 
 
 0.0 
 
 California filago 
 
 California goldenrod 
 
 3.6 
 0.6 
 
 
 2 
 
 Coyote tobacco 
 
 0.0 
 0.0 
 
 Dwarf Bridges grassnut 
 
 0.0 
 
 English plantain 
 
 Foothill death camas 
 
 Fremont death camas 
 
 0.0 
 
 0.5 
 1.2 
 
 
 
 0.0 
 
 
 0.0 
 
 Purple nightshade 
 
 1.2 
 
 9 
 
 
 0.4 
 
 
 0.6 
 
 
 1.6 
 
 
 2.7 
 
 Summer cottonweed 
 
 0.3 
 
 Tarweed 
 
 3.0 
 
 Turkey-mullein 
 
 0.7 
 
 Valley tassels 
 
 0.0 
 
 
 0.3 
 
 Wild carrot 
 
 2.9 
 
 
 5.2 
 
 
 
 Total 
 
 13 5 
 
 88.8 
 
 96.8 
 
 56.7 
 
 28.3 
 
 
 
 Grasses 
 
 
 0.0 
 0.6 
 6.8 
 0.3 
 3.1 
 6.4 
 0.2 
 2.9 
 2.3 
 1.8 
 2.6 
 1.2 
 0.0 
 0.1 
 
 28.3 
 
 2.1 
 
 0.8 
 19.3 
 
 0.5 
 7.7 
 
 21.8 
 7 
 6.7 
 8.4 
 6.5 
 
 14.7 
 6.3 
 2.8 
 0.9 
 
 3.6 
 1.8 
 
 39.1 
 1.4 
 
 14.3 
 
 38.3 
 0.5 
 5.5 
 
 12.6 
 4.2 
 
 22.6 
 9.2 
 7.3 
 1.3 
 
 0.6 
 
 2.6 
 31.2 
 
 2.8 
 11.2 
 30.7 
 
 5 
 
 7.3 
 16.8 
 
 6 7 
 32.1 
 10.7 
 11.2 
 
 2.0 
 
 0.0 
 
 3.9 
 
 59.6 
 
 3.2 
 
 4.9 
 
 21.7 
 
 0.5 
 
 10.4 
 
 7.2 
 
 3.1 
 
 24.6 
 
 15.4 
 
 24.6 
 
 6.8 
 
 185.9 
 
 0.0 
 
 
 4.7 
 
 Foxtail fescue 
 
 63.8 
 
 
 2.7 
 
 
 1.3 
 
 Nitgrass 
 
 36.3 
 
 Purple needlegrass 
 
 0.5 
 
 Rat-tail fescue 
 
 24.1 
 
 
 9.4 
 
 
 2.8 
 
 
 20 1 
 
 Slender oat 
 
 18.1 
 
 Soft chess 
 
 30.3 
 
 Wild oat 
 
 9.6 
 
 
 
 Total 
 
 99.2 
 
 161.7 
 
 166.4 
 
 233.7 
 
 
 
 Average of 24 plots. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 TABLE 5 — (Continued) 
 
 49 
 
 Species 
 
 Before 
 burning 
 
 After burning 
 
 First 
 year 
 
 Second 
 year 
 
 Third 
 year 
 
 Fourth 
 year 
 
 Fifth 
 year 
 
 Brush seedlings 
 
 
 0.0 
 0.0 
 0.0 
 
 0.2 
 
 1 
 
 1.6 
 1.7 
 0.6 
 0.0 
 1.2 
 2.4 
 2.7 
 
 1.2 
 2.6 
 1.3 
 0.4 
 0.9 
 3.7 
 1.8 
 
 2.0 
 2.1 
 2.8 
 0.3 
 0.6 
 3.5 
 2.2 
 
 2.0 
 1.8 
 1.6 
 0.3 
 0.7 
 3.2 
 2.2 
 
 2.0 
 
 
 1.6 
 
 
 1.6 
 
 
 0.3 
 
 
 0.5 
 
 
 3.2 
 
 
 2.2 
 
 
 
 Total 
 
 0.3 
 
 10.2 
 
 11.9 
 
 13.5 
 
 11.8 
 
 11.4 
 
 growth of brush seedlings, in the third year after burning. This resulted in a 
 plant pattern somewhat similar to that of the sprouting covers previously 
 discussed, except that more grass vegetation was present. 
 
 Another point of difference noted in the nonsprouting cover was the marked 
 fluctuation in density of the herbaceous plants during the three years immedi- 
 ately following burning. About one third of the species declined sharply in 
 density the second year ; but these same species increased again in the third 
 year to a density greater than that of the first year. The decline in the second 
 year and increase in the third probably were influenced directly by climatic 
 conditions. Weather data show that precipitation was far below normal for 
 the first two years after burning. In fact, rainfall in December, 1930, was only 
 11 per cent of normal, the lowest recorded for that month. The temperature, 
 
 Fig. 18. — Heavy initial invasion of turkey-mullein on heavily burned cea- 
 nothus areas. By the third year, grasses and other herbs usually replace the 
 turkey-mullein. Lake County. 
 
50 
 
 University of California — Experiment Station 
 
 j, lg iy —plot of oak-grass association burned in 1934, upon which 
 plant succession and soil erosion were studied. A, Showing removal by 
 burning of low bushes, dead branches, and leaf litter. B and C, photo- 
 graphed in 1935 and 1936, respectively, illustrate the progression of the 
 succession from mixed weeds and grass to a nearly pure grass cover. 
 Note recovery of live-oak sprouts, indicated by arrows. Shasta County. 
 
Bul. G85] Plant Succession on Burned Chaparral Lands 51 
 
 while fluctuating greatly, was well above normal during the first two years. 
 Thus, apparently, only those seedlings that could withstand the high tempera- 
 tures and low precipitation showed an increase during this time. The same 
 tendency during the same years was evident, though less striking, in the other 
 nonsprouting areas measured. The adverse climatic conditions appeared to 
 affect little the growth of the sprouting forms, probably because of their rela- 
 tively deep roots. 
 
 In this character of cover, chaparral seedlings were also found in moderate 
 abundance after burning. They tended to occur, however, in rather small but 
 dense patches, leaving much of the area to be occupied exclusively by herb- 
 aceous plants, of which turkey-mullein was among the first to invade, particu- 
 larly on areas with abundant ash (fig. 18) . The increased carrying capacity on 
 the deeper, more level soils of this cover was much improved by burning. 
 
 Cover of Interior Live Oak and Blue Oak. — Since much of this rather exten- 
 sive association, dominated by interior live oak and blue oak, common to the 
 lower stretches of the foothill region, is readily accessible for pasture use, the 
 brush and tree growth is often greatly altered by some form of clearing other 
 than direct burning. (See p. 108 and 111.) The soils of most such areas are 
 fairly deep and productive, and sprout and other reproduction come in 
 promptly and vigorously following a fire. If permitted to develop undisturbed 
 after burning, the sprouts readily suppress the understory vegetation, and 
 often recapture the area almost to the exclusion of other vegetation. 
 
 The study of plant succession of the sprouting-oak cover was intensively 
 undertaken in Shasta County in 1931, and was later supplemented by check 
 plots in Tehama, Butte, Mendocino, Lake, El Dorado, and Alameda counties. 
 
 The Shasta County plots were located on rather level lands east and north 
 of Redding, in an important livestock-producing area. Typical of this locality, 
 the dominant cover consisted of interior live oak and blue oak, with subdomi- 
 nants of common manzanita, whiteleaf manzanita, buckbrush, madrone, and 
 poison oak (fig. 19). Much the same association of brush plants occurs exten- 
 sively in the counties named above, where check records were made. The 
 soils, mostly of the Redding gravelly loam and Bellavista sandy loam types, 
 varied from shallow to deep, and were of fair to good quality. The plant suc- 
 cessional behavior is presented in table 6. 
 
 The table reveals a fairly large and diversified flora of broad-leaved herbs 
 before and after burning, although before burning the understory vegetation 
 was relatively sparse. Probably because of the tendency of the subdominant 
 brush to f oral a more complete or nearly closed canopy, 5 of the 21 species of 
 broad-leaved herbs listed were not in evidence on the 32 plots before burning ; 
 but they were found widely scattered over the experimental area as a whole. 
 The largest population of broad-leaved herbs, fourteen times that of the 
 unburned plots, appeared the first year after burning, following which there 
 was a gradual decline as the sprouting brush reclaimed the area. Thus in the 
 fifth year, there were barely three times more herbaceous plants than before 
 burning, in 1931. The most conspicuous broad-leaved herbs in these plots, and 
 in this cover in other counties, were turkey-mullein, Napa star thistle, willow 
 herb, Spanish-clover, and the poisonous coyote tobacco, with subdominants of 
 slender buckwheat, tomcat clover, and rattlesnake weed. The two first-named 
 
52 
 
 University of California — Experiment Station 
 
 TABLE 6 
 
 Average Number of Understory Plants per Milacre Plot in Live Oak and Blue Oak 
 
 Cover, before Burning in 1931, and for Each of the Five Years 
 
 Following* ; Shasta County 
 
 Species 
 
 Before 
 burning 
 
 After burning 
 
 First 
 year 
 
 Second 
 year 
 
 Third 
 year 
 
 Fourth 
 year 
 
 Fifth 
 year 
 
 Broad-leaved herbs 
 
 Bush beardtongue 
 
 Cantua spurge 
 
 Chaparral cottonweed 
 
 Common soap-plant 
 
 Common vervain 
 
 Common yellow mustard . 
 
 Cotton-batting plant 
 
 Coyote tobacco 
 
 Hill lotus 
 
 Horseweed 
 
 Indian pink 
 
 Napa star thistle 
 
 Purple nightshade 
 
 Redstem fi laree 
 
 Nada stickleaf 
 
 Slender buckwheat 
 
 Small-flowered lotus 
 
 Spanish-clover 
 
 Turkey -mullein 
 
 White-flowered navarretia 
 Willow herb 
 
 Total 
 
 0.2 
 3 
 01 
 0.1 
 
 2 
 1 
 1 
 1 
 
 
 3 
 
 5 2 
 
 1 2 
 
 5 1 
 
 4.6 
 
 14 
 
 3 7 
 
 2.6 
 
 2.0 
 
 3.9 
 
 1.6 
 
 1.9 
 
 1 
 
 9 
 
 1 
 
 2 
 
 4 
 
 3.6 
 
 1.7 
 
 5.8 
 
 3 5 
 
 6.2 
 
 3.8 
 
 72 3 
 
 0.7 
 2 8 
 2.9 
 1.3 
 1.2 
 2.3 
 1.6 
 4.2 
 1.8 
 
 2 4 
 12 
 5.8 
 13 
 2.7 
 3.2 
 2.4 
 2.8 
 3.9 
 3.8 
 3.7 
 
 3 3 
 
 55 3 
 
 9 
 13 
 
 5.2 
 7 
 8 
 14 
 8 
 1.7 
 1.2 
 1.3 
 
 45.1 
 
 
 
 
 2. 
 
 
 
 
 1 
 1. 
 
 8 
 3 
 7 
 3 
 3 4 
 1.1 
 1.1 
 1.2 
 1.4 
 7 
 2.3 
 2.1 
 1.7 
 1.6 
 
 25 
 
 0.5 
 4 
 1.6 
 4 
 0.4 
 1.6 
 0.5 
 3 
 1 
 2 
 0.1 
 1.3 
 6 
 6 
 3 
 6 
 3 
 14 
 17 
 1.2 
 0.8 
 
 
 0.0 
 
 4.2 
 1.3 
 0.0 
 3.6 
 0.3 
 0.4 
 8.3 
 15 
 0.3 
 
 1.2 
 
 16.7 
 
 10.2 
 7 
 
 19.2 
 4 4 
 7.6 
 
 29.2 
 6.6 
 2.9 
 
 2 4 
 11.8 
 4 3 
 19 
 11.4 
 5.2 
 3.9 
 14.4 
 6.7 
 3.1 
 
 4 
 
 3.6 
 2.2 
 0.3 
 6 3 
 17 
 3.8 
 6.6 
 5.8 
 1.6 
 
 1 
 2.5 
 
 2 4 
 0.0 
 4.8 
 0.8 
 1.7 
 
 11.7 
 
 3 
 0.3 
 
 0.0 
 
 
 2.7 
 
 
 1.1 
 
 
 0.0 
 
 Nitgrass 
 
 2.3 
 
 14 
 
 
 14 
 
 
 9.3 
 
 
 1.3 
 
 Wild oat 
 
 0.6 
 
 
 
 Total 
 
 19.9 
 
 98.7 
 
 65 1 
 
 32.3 
 
 27.3 
 
 20.1 
 
 
 
 Brush seedlings 
 
 Chaparral coffeeberry 
 
 0.0 
 0.0 
 0.0 
 
 0.0 
 
 1.2 
 2.2 
 0.0 
 0.4 
 3.4 
 
 1.4 
 2 4 
 0.2 
 0.3 
 2.3 
 
 1.1 
 2.0 
 0.2 
 
 3 
 
 2.1 
 
 5.7 
 
 6 
 13 
 0.2 
 0.2 
 2 1 
 
 6 
 
 13 
 
 
 0.2 
 
 
 0.2 
 
 Whiteleaf manzanita 
 
 2.1 
 
 
 
 Total 
 
 0.0 
 
 7.2 
 
 6.6 
 
 4.4 
 
 4 4 
 
 
 
 'Average of 32 plots. 
 
Bul. G85] 
 
 Plant Succession on Burned Chaparral Lands 
 
 53 
 
 species often are the exclusive occupants of spots where much ash has accu- 
 mulated. On such spots, for three or more years, these plants frequently com- 
 pose most of the cover (fig. 20) . 
 
 Although the grass vegetation consisted exclusively of annual species on the 
 plots in Shasta County, a few perennials occurred on some of the sample areas 
 
 
 
 ... . • • . 
 
 . . . ...,■■.■ . . . . ■ ■ ■ ■ ■. .. ■ . . ..... 
 
 
 
 
 
 " : " : """ 
 
 ■."'.' ^s^P f< ?*^"^s9Kx^ flaMO 
 
 r*v' V .'' V^ 5 
 
 ftjjfh' 
 
 
 
 MbH 
 
 Fig. 20. — .4, The dotted line indicates the heavy accumulation of ash left 
 from burning of a dense stand of manzanita; turkey-mullein is the first in- 
 vader. B, Napa star thistle encircling and encroaching upon the ground the 
 following year, as a secondary invader. In the third or fourth year grasses 
 and other herbs also appear. Lake County. 
 
 in other counties, purple needlegrass being the most common and widespread. 
 Soft chess, one of the most valuable of the annual grasses, was well represented 
 for four years after burning. As in the other covers reported, however, the 
 early-maturing and less robust fescues, silver hairgrass, and nitgrass were 
 numerically the most abundant. By the fifth year after burning, the invading 
 brush sprouts and seedlings had nearly choked out the grasses and broad- 
 leaved herbs. 
 
54 
 
 University of California— Experiment Station 
 
 Despite the presence of the dense, vigorous sprouting, the brush stand was 
 further rejuvenated by the establishment of seedlings of the various brush 
 species which were abundant before burning. More than half of the brush 
 seedling stand recorded the first year after burning appeared to have promise 
 of permanence in the fifth year. Broadcast burning usually resulted in con- 
 siderable increase in carrying capacity, but this increase was of short duration 
 unless followed with measures to suppress the aggressive brush. 
 
 Comparisons of the succession of grasses, broad-leaved herbs, and brush 
 seedlings recorded on all the plots of the sprouting and nonsprouting chapar- 
 ral areas are shown in figure 21. The number of grass plants, although com- 
 
 2 3 
 
 YEARS AFTER BURNING 
 
 Fig. 21. — Average number of herbaceous plants and brush seed- 
 lings per milacre on sprouting and on nonsprouting covers, before 
 and after burning. 
 
 posed largely of the early-maturing, low-yielding foxtail fescue, is seen to 
 have increased in general on the plots of sprouting brush for the first year 
 after burning. In the second year the grass stand increases only slightly on 
 the sprouting chaparral plots, and thereafter the grasses typically decline in 
 number. In contrast, on the plots of the nonsprouting chaparral lands the 
 grasses continue to increase for each of the five years after burning, and the 
 more robust and valuable forage grasses gradually replace the inferior foxtail 
 fescue. The successional population of broad-leaved herbs (weeds) on the 
 sprouting and nonsprouting brushlands is of the same general pattern, being 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 55 
 
 highest in the second year, and then rapidly declining. The brush seedlings are 
 most numerous in the plots of sprouting chaparral, the maximum number 
 occurring the first year after burning. Competition with the abundant, vigor- 
 ous crown sprouts and seedlings of the sprouting chaparral accounts chiefly 
 for the decline in grasses and broad-leaved herbs of that cover, whereas the 
 
 Fig. 22. — A, Oak and grass cover of good grazing capacity which should not 
 be broadcast-burned if chamise is in or near the area. Kemoval of the trees by 
 girdling or chopping would improve the land for grazing. B, Chamise seed- 
 lings, recognized by the dark stems in the grass cover, show extent of inva- 
 sion from a single chamise bush following burning; the mother plant is shown 
 at the left. El Dorado County. 
 
 increasing abundance of grass during the five-year cycle, and to a lesser extent 
 of the invading brush seedlings, evidently accounts for the decline in the 
 broad-leaved herbs on the nonsprouting areas. In sites of thin soil, especially 
 those located on steep south and west slopes, there are conspicuously limited 
 invasions of grasses and other herbs. North and east slopes of similar brush 
 types, on the other hand, produce considerably more forage following burning 
 than do the drier south and west slopes. 
 
56 University of California — Experiment Station 
 
 WIDENING OF CHAPARRAL AREAS BY BURNING 
 
 A statement often made by those who own foothill lands is that the brush 
 fields now occupy greater acreage than formerly. This contention has been 
 supported by several workers. Kotok (55), for example, has pointed out that 
 in the Sierra Nevada foothills sweeping changes have taken place in the forest 
 in the past fifty to seventy -five years as a result of fire. After fire, or log- 
 ging and fire, pure ponderosa pine lands have largely passed to a deerbrush 
 and whiteleaf manzanita association, and the warmer slopes have become a 
 thicket of interior live oak. In other areas where severe fires have been fre- 
 quent, the pine lands have given way first to a manzanita-chamise cover, and 
 ultimately to pure chamise. Moreover, on grassland bordered by areas of chap- 
 arral, as in El Dorado County, fire has resulted in extending the chaparral into 
 the grassland (fig. 22). According to Weeks, Wieslander, and Hill (114), 
 "burned-over chaparral areas indicate that brush fields are extended as a 
 result of fire at the expense of grass." 
 
 The plot technique already discussed was used on a number of grasslands 
 bordering chaparral in Mendocino, Lake, and Shasta counties. The aim was to 
 note the degree and distance of extension of the brush into the adjoining grass- 
 land. Plots were established only where one-year-old brush seedlings had come 
 in. Several such plots were mapped yearly to ascertain the survival percentage 
 of the invading brush seedlings. These small-plot measurements were supple- 
 mented by extensive observations and approximate measurements to verify 
 the more restricted quantitative records. 
 
 Although the findings with respect to brush invasion varied, extension of 
 chamise, and various species of manzanita and ceanothus, into the burned 
 brush-bordered grassland was found to occur with rather marked regularity. 
 On the side of the mature brush areas leeward to the prevailing winds, scat- 
 tered brush seedlings occasionally extended as much as 50 yards into the 
 burned grassland, whereas at distances of 5 to 15 yards from the brush border 
 they were much more abundant, varying from 1 to as many as 15 seedlings per 
 milacre. On the windward side of the brush field, chaparral seedlings were 
 considerably more restricted, both as to numbers and distance from the parent 
 stand, except occasionally along the steeper slopes where the seed had presum- 
 ably been carried farther from the adult plants by force of gravity. 
 
 Survival of the invading seedlings varied from a total loss, as where the 
 growth of grass was heavy, to complete establishment where grass competition 
 was weak. The survival percentage of the one-year-old plants doubtless was 
 also greatly influenced by the climate of that season. On an average, slightly 
 more than 50 per cent of the invaders survived at the end of the first year, and 
 only about 15 per cent of the remainder died thereafter. This more or less 
 dense advance growth appeared to serve as a potential source for further 
 extension of the brush during the eight-year to ten-year interval between 
 burnings. During this interval, seed crops are accumulated in the soil, and 
 further extended into the grassland. A small proportion of these may germi- 
 nate without the stimulus of fire, but the majority remain dormant until a fire 
 revisits the area, after which they germinate vigorously. 
 
 One of the most impressive cases of brush advancement following burning 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 was that of a 3-acre community of chamise in the steeper part of an oak wood- 
 land cover of Lake County (fig. 23). Superficially, the stand gave the appear- 
 ance of uniformity in height and age, except for a bordering circle of young, 
 low-growing chamise plants; on the uppermost edge of the stand these ex- 
 
 Fig. 23. — A, Chamise area, outlined by dotted line, containing three units 
 of even-aged stands evidently resulting from fires, as determined by growth- 
 ring counts and fire scars of crowns. B, Chamise and wedgeleaf ceanothus 
 advancing into formerly open grassy oak woodland after one fire. Another 
 fire is likely to result in a much denser stand of the brush. Lake County. 
 
 tended only a few yards, but laterally and beyond the lowermost edge of the 
 mature stand, they extended from 15 to 50 yards in width. 
 
 In an attempt to establish age of isolated stands of chamise and to note any 
 interruptions in its development, growth-ring counts were made of the root 
 crowns on milacre sample plots. The plots were spaced at intervals of 20 feet 
 
58 
 
 University of California — Experiment Station 
 
 through the stand referred to above, starting at the uppermost portion of the 
 young stand and ending at the lowermost edge of the chamise community, and 
 also traversing the area laterally at two well-spaced intervals. These ring 
 counts showed that the mature portion of the stand was composed of at least 
 
 Fig. 24. — Cross sections of chamise crowns taken from the oldest portion of the area 
 illustrated in figure 23. The crowns in A, B, and C show asymmetrical widths resulting 
 from fire ; D shows symmetrical development of a crown little damaged by fire. Because of 
 growth distortions on burned areas, the maximum age attained by chamise and other such 
 crown-producing species cannot always be determined accurately. 
 
 two distinct age groups, even though the brush was approximately uniform in 
 height. Occurring as an irregular patch through the central portion, and 
 occupying about one fifth of the total area of the stand, were chamise plants 
 whose root crowns averaged 22 growth rings, varying from 18 to 24 (fig. 24). 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 59 
 
 The development of nearly all the root crowns within the central area was 
 asymmetrical, most of the rings having been formed on the side opposite the 
 healed-over but discolored fire scars. The remainder of the even-height growth, 
 with its fully developed crowns, composing approximately three fifths of the 
 stand, contained bushes with root crowns averaging 11 growth rings, the vari- 
 ation being 10 to 12. The root crowns of this growth were relatively more 
 symmetrical, although most of them were fire-scarred. The crowns of the 
 young chamise plants, located on the periphery of the older stand, had 4 
 growth rings, but showed no fire scars. Thus, the central portion of the stand 
 showed an average age of twenty-two years, and the effects of at least two 
 heavy fires. Surrounding this, the greater portion of the area averaged eleven 
 years, and showed the effects of moderate burning by one heavy fire. The outer- 
 most circle of young brush showed no effects of burning, indicating that the 
 last fire had crossed the area about five years previously. The extension bound- 
 ary of the brush area after each fire was clearly indicated in the ring counts 
 of the root crowns. 
 
 This plot and ring-count study, supplemented by extensive observations 
 and ring counts made in many localities of the northern counties, supports the 
 conclusion of other workers that burning at infrequent intervals may, and 
 often does, extend the chaparral cover. The heat generated by a brush fire, as 
 shown by laboratory tests (p. 61), activates the seeds of some chaparral 
 species which apparently lie dormant for long periods under the unburned 
 ground cover (fig. 23, B). Moreover, the root crowns of sprouting brush species 
 are remarkably adjusted to endure severe and frequent burning. Unless the 
 entire periphery of the root crowns is deeply charred, as seldom occurs, 
 subsequent vigorous sprouting takes place after a fire. The precautions of 
 operators to keep fire out of areas of scattered brush are justified by the studies 
 here reported. 
 
 REPLACEMENT OF CHAPARRAL BY GRASS THROUGH 
 EXCLUSION OF FIRE 
 
 Although chaparral is evidently the most permanent and persistent vege- 
 tation occupying extensive foothill areas of the southernmost counties of the 
 state, as well as many of the more inferior sites in the northern counties, it 
 also occurs sometimes as a subclimax, or more temporary cover, on sites favor- 
 able to the growth of other vegetation. The fact that fires often result in 
 thickening the existing chaparral stands, and in extending the brush into 
 bordering lands, does not necessarily imply that total exclusion of fire would 
 reverse the process and result in replacement of the brush by herbaceous 
 vegetation. 
 
 A few chaparral areas have been examined which, according to fire scars, 
 ring counts of the root crown, and historic records, had not been burned for 
 more than half a century. Even with this long period of nonburning, shallow 
 and poor soils continue to support brush which forms nearly a complete cover, 
 with no tree growth, and with little competing understory vegetation. Such 
 unburned stands are largely composed of uneven-aged plants, indicating that 
 as clumps of chaparral naturally opened up or died out, an occasional hard- 
 brush seedling would survive, its establishment being favored by the declin- 
 
60 University of California — Experiment Station 
 
 ing competition of the decadent brush. Most old unburned chaparral stands, 
 however, become decadent within a relatively short time. Branches of old 
 chamise plants, for example, as judged by their growth rings, die off or be- 
 come virtually defoliated at twenty to twenty-five years of age. After this 
 decadent stage, the brush areas which happen to occupy the better soils are 
 strongly reclaimed by grass or forest, according to which form of vegetation 
 was climax. In such replacement, the few brush seedlings which appear simul- 
 taneously with the opening of the chaparral cover are suppressed and eventu- 
 ally eliminated by competition with the invading climax vegetation. Beneath 
 the brush cover, on sites of good soil, a fairly dense, vigorous stand of purple 
 needlegrass and various herbs is frequently found, the roots of which occur 
 in abundance throughout the upper foot or so of soil. Luxuriant grass growth 
 also occupies open spaces between the brush plants, giving the grass vegeta- 
 tion an appearance of dominance over the chaparral. 
 
 Where the replacement of brush by grass is so far advanced that the under- 
 story vegetation occupies 20 to 30 per cent of the ground under a combination 
 of dead and growing branches of the brush, burning is almost certain to result 
 in reversion of the vegetation of the area to its former tangle of brush. A note- 
 worthy instance of this was observed over a period of several years in Mendo- 
 cino County. Approximately 100 acres of a mature, gradually opening stand 
 of chamise was being invaded by a heavy needlegrass cover. Despite the fact 
 that sheep were able easily to forage over most of the tract, the area was 
 heavily burned. The profusion of vigorous chamise sprouts and seedlings 
 which followed the fire so greatly suppressed the grass vegetation after the 
 second year as to render the area useless for grazing for several years there- 
 after. Had fire been kept out, this area would likely have become predomi- 
 nantly or wholly a grass cover, similar to the adjoining lands which supported 
 a fair amount of bunch-grass vegetation. 
 
 In brief, the evidence strongly indicates that the less favorable sites, now 
 almost completely occupied by chaparral and chamise, have supported, and 
 will continue to support, the brush cover for indefinite periods. Neither burn- 
 ing nor the exclusion of fire is apt to affect measurably the character or 
 abundance of the brush occurring on thin or otherwise inferior soils. On sites 
 of high quality on the other hand, exclusion of fire will in many instances 
 result in replacement of the brush by the true climax vegetation. Thus, where 
 the brush is thinning out because of decadence and is being replaced by 
 grasses, especially perennial forms, fire should be kept out. If, however, the 
 brush cover on productive lands is vigorous, the period of fire exclusion 
 necessary to induce an abundant growth of grass should be weighed against 
 the cost and hazards of brush burning in adopting a management plan. Fire- 
 exclusion periods of twenty years or more may have to elapse before the brush 
 on most good lands will thin out sufficiently to favor the growth of an abun- 
 dant understory vegetation. Such extended fire-protection periods are not 
 always economically practical, especially if many of the intervening years 
 are marked by low forage values. Recognition of the turning point in a brush- 
 and-grass struggle for successional supremacy is, notwithstanding, of great 
 importance in a brush -suppression program on areas that have not been 
 recently burned. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 61 
 
 SOIL TEMPERATURES DURING BURNING 
 
 The heat of fire has an obvious effect in killing the leaves and branches of 
 plants ; but its effects on the chemistry of the soil, and especially on germina- 
 tion of seeds lying in the soil litter, are also important influences in affecting 
 the plant cover. This view is widely accepted and has been verified by several 
 research workers. 
 
 Hey ward (37) measured soil temperatures during fire in longleaf pine for- 
 ests. Hofman (42), working in the Douglas-fir region, and Korstian (54), in 
 deciduous forests, also studied soil temperatures during fires. Results from 
 forest fires are not directly applicable to brushland, however, and information 
 concerning the actual magnitude and duration of high temperatures is lack- 
 ing. For this reason it seemed desirable to obtain data as to the heat created in 
 the litter and surface soil during the burning of different amounts of brush 
 fuel. The data here presented include also the more pertinent results of studies 
 by Craddock. 21 
 
 The field fire temperatures were measured by a recording Bristol pyrometer 
 of special design. 22 After selection of a site for study, the thermocouples were 
 inserted in the ground. In recording most of the readings, the hot junction of 
 one thermocouple was placed % inch deep in the litter ; the second and third 
 junctions % and l 1 /*? inches deep, respectively, in the soil proper. Where tem- 
 peratures were procured under chaparral, the tips of the thermocouples were 
 placed midway between the stem bases and the periphery of the crown radius; 
 and for readings in the grass cover the tips were located outside of the crown 
 periphery. Temperatures were recorded at 1-minute intervals for each thermo- 
 couple until the fire had subsided. 
 
 The pyrometer fire data are presented in table 7. These data show that the 
 time required to reach the maximum temperature is relatively short in the 
 litter and duff compared with that in the surface soil. In the 11 field brush 
 fires studied the maximum temperature in the litter was reached in 1 to 9 
 minutes, whereas at depths of % to 2 inches in the soil proper the time required 
 to reach maximum temperatures, with one exception, varied from 7 to 31 
 minutes. The duration of temperatures above 150° F in the soil was directly 
 proportional to the duration and the intensity of the temperatures at the sur- 
 face, as shown in figures 25, 26, and 27. Figures 25 and 26 bring out the 
 characteristic sharp rise in temperature in a heavy litter, with abundant burn- 
 ing fuel above, where the maximum temperature' was reached in the litter in 
 less than 10 minutes. Figure 27 shows the more gradual rise to the maximum 
 temperature in the litter where less fuel is available. The lag in reaching the 
 maximum temperature in the deeper soil layer, as shown in curve C of figure 
 25, is characteristic of all such readings. 
 
 21 George W. Craddock, working under the direction of the writer at the University of 
 California, in 1927 presented data on soil temperatures during chaparral fires in his thesis, 
 "The Successional Influence of Fire in the Chaparral Type." This is on file in the University 
 of California Library, Berkeley, California. 
 
 22 In checking the instrument against a standard potentiometer, the readings were found 
 to be in agreement within 5 degrees in the range of 100° to 500° F, and within 10 degrees 
 above the 500° range. The sensitive pendulum marker responded accurately at 10-second 
 intervals to wide fluctuations in temperatures as transmitted from the thermocouples when 
 variously situated with respect to the heat factor. 
 
a O ea —< 
 .5 « g u 
 
 CM Oi CM rH 
 
 S82 
 
 ,-< .-I ^(ItOW 
 
 op 
 
 n as 
 
 3 
 
 o 3-g 
 
 
 S-a *> 
 
 
 
 £ 
 
 
 a 
 
 3* 
 
 a 
 1 
 
 co <o u 
 
 ft 60 
 
 s 
 
 a^ 
 
 i co •»»< t-co 
 
 i-i rt(N 
 
 mm >o O 
 
 lOCD CD ^h 
 
 ooo ooo «coo ©o»o 
 
 OMH 05<M-^< MNCC -^ r-* CO 
 <M CO CO OS ^f CM ©CON QO •<!«< CM 
 
 J) 
 
 fto « 
 
 4) fl C 
 
 fig'" 
 
 d.a 
 
 
 T, a- — -, s u 
 ."E 3 *C O co 
 
 ;£ so o ."£ so -5 
 
 S .2 .2 c c '~ 
 
 *» "» • 0~ -3o rr,— . CJ 
 
 tn O J^. O ii O co 
 
 _ fcl 73 SO _ 77J <*> _ 
 
 fic-2 o_ s .2 
 
 to-i*]"* « . . 
 = *S8 s^-a 
 
 73 co _. "OOO 
 
 all 
 
 •5*1 -5 c c -5*1 -3*1 .5 '1-5 .5"! .5*1.5 .5.51 B.5'1 .5.5*1 -Sec 
 
 ^ $£ 2$5 ^s ^&? ^^ ^§^ ^s o^ ^s ^55 
 
 >+J a 
 
 73 u 
 
 •I a 
 
 ft M 
 
 
 5 -o 
 
 CO « 
 
 ■s IS 
 
 3 5 
 
 co o 
 
 si 
 
 c3 <o 
 
 "ST"* 
 
 -si 
 
 73 c3 
 Oi O 
 £ (1) 
 
 2.2 
 
 C3 ai 
 
 73 e3 
 .2t3 
 
 .-.2 o 
 
Bul. C85] Plant Succession on Burned Chaparral Lands 
 
 1,000 
 
 G3 
 
 h 
 W 
 
 X 
 
 Z 6OO 
 
 w 
 a 
 
 X 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -—A 
 
 % INCH, IN LITTER 
 
 1 ■*- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 , % INCH, IN SOIL 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 7^ 
 
 / 
 
 / 
 
 
 \ 
 
 
 
 
 
 
 
 / 
 1 
 1 
 1 
 
 
 
 
 ^ 
 
 '*''•»» 
 
 
 
 
 
 1 
 
 1 
 
 1 . 
 1 / 
 
 7^ 
 
 ,,lJ£ INCHES.1N SOIL 
 
 
 "*"* 
 
 ^^5 
 
 ^^ 
 
 
 !/ 
 
 1/ 
 
 
 
 
 
 
 
 
 
 
 20 30 AO 50 6O 70 80 90 
 
 MINUTES 
 
 Fig. 25. — Soil temperatures recorded during chamise fire num- 
 ber 8: A, Thermocouple, % inch in duff; B, % inch in the soil; 
 C,1V2 inches in the soil. Mendocino County. 
 
 £ 6 °° 
 w 
 X 
 2 
 
 W 
 
 a 
 
 £ AOO 
 
 w 
 
 w 
 
 w 
 p 
 
 
 
 
 
 
 
 
 
 
 
 f 
 
 ^x 
 
 V 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 s^ 
 
 Yz INCH , IN LITTER 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 " 1 INCH, IN SOIL 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 AO 50 60 70 SO 
 
 MINUTES 
 
 Fig. 26. — Soil temperature recorded during a fire in mixed 
 chaparral and grass. Sonora County. (After Craddock.) 
 
G4 
 
 University of California — Experiment Station 
 
 Probably the most significant point brought out in these records is the dura- 
 tion of the heat in the surface soil where most of the seeds have lodged. Tem- 
 peratures above those endured by most seeds were recorded in several in- 
 stances, notably in experimental field fire stations 1, 7, 8, 10, and 11, shown in 
 table 7. This fact was verified by subjecting to conditions of germination the 
 seed contained in the burned soil collected at the depths to which the thermo- 
 couples had been set, and comparing the results with those of corresponding 
 soil samples taken on adjoining unburned spots. Transfer of these samples to 
 fiats in the greenhouse, where they were watered at suitable intervals, revealed 
 that the seed of nearly all species had been killed where the soil temperature 
 was maintained for several minutes above 300° F. 
 
 The soil-temperature data here presented for individual fires may not be 
 wholly representative of the temperatures created when a large brush field is 
 
 14 
 
 X 400 
 
 U 
 
 a 
 X 
 
 < 300 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 v 
 
 ^ IN GRASS AND LITTER 
 
 
 
 
 
 
 
 
 
 \ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 ( 
 
 ,. 1 INCH.1N SOIL 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 2 4 6 8 10 12 14 16 18 20 22 24 26 28 
 
 MINUTES 
 
 Fig. 27. — Soil temperatures recorded during a fire in a manzanita 
 and grass cover. The litter layer was thin. Humboldt County. 
 
 burned, yet the trends of the data appear to conform to specific patterns. Fig- 
 ures 25, 26, and 27, for example, show irregularities in the temperature curves 
 for the litter which are purely local. The falling of burning twigs and 
 branches, the exact placing of the thermocouples in relation to the fuel, the 
 kind of fuel, the extent to which the fuel was consumed, wind movement, rela- 
 tive air humidity, and many other related factors influence to some extent the 
 "spot" temperature records. The hottest fires were recorded in slash, the mixed 
 chaparral coming second, chamise next, and the grass last. The fact that 
 surface-soil temperatures within the crown circumference of the larger indi- 
 vidual bushes frequently exceeded the viability endurance of the accumulated 
 seed appears partly to account for the temporary barrenness of small areas 
 within larger burns. The toxicity of heavy accumulations of ash, however, 
 must not be overlooked as a factor adverse both to the germination of seeds and 
 to the establishment of seedlings. 
 
 HEAT STUDIES WITH SEEDS 
 
 Evidence supports the belief that seeds of the hard sclerophyll species, like 
 those of many other kinds of cover, may lie dormant, but retain their viability 
 for many years {23). In some species the after-ripening process may last for 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 65 
 
 considerable time regardless of the treatment of the seed ; in others the integu- 
 ments may be so thick or of such dense structure as to preclude oxidation and 
 the absorption of moisture until some factor alters the seed coat (8, 23) . Heat 
 treatment is one factor which may hasten germination of the seeds. Moreover, 
 conditions to which seeds are subjected after being heated may affect their 
 viability and rate of germination (78) . 
 
 Where sufficient fuel is available to kindle a particularly hot fire, a "selec- 
 tive" seedling stand appears to come in, with the result that certain hard 
 sclerophylls become chief occupants of the site. This apparent selectivity of 
 species induced the writer to initiate a study of the effect of heat on the ger- 
 mination of several of the more common chaparral and associated herbaceous 
 species. The more pertinent results of several carefully supervised researches 
 of Wright (117) are included with the data here presented. The chief aim was 
 to note the relative high-temperature endurance of the seeds of various chap- 
 arral and grass species, with a view to explaining why certain plants com- 
 monly reclaim burned areas. 
 
 The shrub and grass seeds used in the laboratory heat treatments were 
 collected in the chaparral belt from the current crop of four different years. 
 The shrub species tested were 23 coffeeberry, California snow-drop bush, 
 chamise, chaparral whitethorn, common manzanita, hoarjdeaf ceanothus, lau- 
 rel sumac, Parry manzanita, sugarbush sumac, and western chokecherry. The 
 grasses tested were purple needlegrass, ripgut, blue wild-rye, soft chess, and 
 wild oat. All seed collections were air-dried and freed of capsules, glumes, and 
 other such covering before being tested. They were stored in a dry, cool room 
 until treated. 
 
 Artificial heating of the seeds was done in lots of 50 to 100 of each species, 
 and 100 unheated seeds were used in the check lots. The seeds were exposed 
 exclusively to dry heat. A thermostatically controlled electric oven was em- 
 ployed in heating the seed. 
 
 In order to ascertain the temperature endurance of the different species of 
 seeds used, the several lots were exposed for 5-minute periods to temperatures 
 ranging from 120° to 300° F, at intervals of 20 degrees. After heating, both 
 the treated and the check (unheated) seed lots were placed between moist blot- 
 ters in an electric germination oven, where they were maintained at a tem- 
 perature of 68°. In addition, several lots of both were germinated in sand beds 
 in a greenhouse, as a further check. Tap water was applied daily to keep the 
 medium moist. Germination counts were made weekly. The germination data 
 of the several common chaparral and grass species are given in table 8. The 
 table shows that the seeds of 14 of the 21 species of shrubs studied gave con- 
 sistently higher germination percentages after being heated than did those of 
 the checks, or unheated seeds. Samples numbered 7, 10, 11, 12, 14, 15, 17, 18, 
 19, 20, and 21, which represent seeds having various forms of integuments and 
 densities of the testa, gave much higher germination percentages after heating. 
 The seeds of several of the other species reacted slightly in favor of the heat 
 treatment, whereas the germination of some was not affected by exposure to 
 the temperatures used. Studies of germination under field conditions have 
 
 - 3 The several species that failed entirely to germinate with or without heat treatment are 
 not here listed. 
 

 m^ 
 
 
 J£ 01 
 
 
 
 
 
 
 
 
 3 h 
 
 -* 
 
 
 Sfe 
 
 
 -fj 
 
 CO O 
 
 
 
 
 
 13 
 
 
 a, 
 
 -Sfc 
 
 
 
 
 oo 
 
 
 
 3 
 
 CM <M 
 
 
 
 co 
 
 S* 
 
 ft 
 
 o o 
 
 a 
 
 5§ 
 
 
 CM CM 
 
 2* 
 
 
 
 o 
 
 ©o 
 
 
 
 &0 
 
 NM 
 
 3 
 
 
 
 "c3 
 
 3En 
 
 
 
 A 
 
 <=>p 
 
 
 
 o 
 
 CM^ 
 
 
 
 03 
 
 5^ 
 
 3 
 
 
 C 
 
 oo 
 
 
 
 03 
 
 
 
 
 s 
 
 Sfc 
 
 
 
 CD 
 
 o© 
 
 M 
 
 
 3 
 
 "- 1 '""' 
 
 
 o 
 
 Sfr 
 
 
 
 
 oo 
 
 Pk 
 
 ^CO 
 
 
 
 3^ 
 
 
 
 
 oo 
 
 
 
 
 
 3fc 
 
 
 
 
 «=>o 
 
 
 
 
 "H-* 
 
 09 
 
 
 
 8 
 
 a 
 
 m 
 
 
 a 
 
 
 > 
 
 >> 
 
 
 3 
 
 3 
 
 O 
 
 O 
 
 
 11 § 
 
 
 CO 
 
 3> i - /. f ic o a -^ ^ ci ic cc -r cc x m m n o Tf o 
 
 OOOOOOMM'J'OOINOMMOOOOOO 
 
 OO'-NON0)NIM00tU5O(DMOt0N(MCCN 
 
 ^ r-IN(N(NMCO 
 
 © © © Q5 O l~ — HC00 00O1hO00hOICO1!BN[- 
 — rtMNHffll^H NM-"-< CM CM — 
 
 NP3 -h rt im IN -- CO m iCO -< CM CM — i 
 
 OM^MOOI-r-C-lrOtKMiM-fOaNOfW' 
 
 r^ co ~ ^- ■— cm cm — < co •*»< ^h ^h 
 
 *N»ujN«iNna:^NO»0^»OOOi 
 
 tOCO l—l 1—1 i-l -^ r^ .-I 1— I CM i-H r-i ,-1 
 
 i CM CO t- CO CM — 'TfcOCTticMOMCMOiOeMfrOCM 
 
 *CS1<M -ClmOCOO!5!CO-HO(NfCiO 
 
 £ ~ & c 
 ffllpp 
 
 CD CD i-' U 
 
 cd cd £ £ 
 cCcC O O 
 
 OOflfl 
 
 W CJ CO CO 
 
 r « 
 O O-S 3 
 3 .3 -3.3 
 CD CD Ct3 o 
 
 S34J 2 c 
 
 2 ■ O O 0) 
 
 3 _g c3 c3X! 
 
 5-s S S 8 
 
 q q 5 3 3^ 
 
 
 CO CO V 
 
 3 
 
 ft Is .sla'i e a S.£g "&1TS ll&Va $ 
 
 X^13 ^^^^j-jsi: o O g g.M.SPcl 3 3.* 
 
 lO^ti--' '->"-' ' 
 
 o o 
 
 3 3 
 CO* 
 
 u O O 
 
 pq a> a> 
 
 cq 
 
 CD CD 
 
 3 3 , 
 
 OoJcUOO - -cSOOS 
 3^.3,3^ 4 *£ 3 3 » 3 S3 §^ § c« O q.3 
 
 •HCNcO^mcD^WaO^NM'O'lOCCt-OOOO'^ 
 
 -(-.Hffl^i-onooon* 
 
 CO^rcjiOOCTlCOOOiQOcOiOt^ 
 
 oooooooooooo 
 
 oooooooooooo 
 
 ON'I'HMOOOOOOO 
 
 t»01»M(ONONNONCO 
 
 CM ^< t~ t~ C 
 
 MfOMw^NNNOM-HO 
 
 "^ cm — oo co r- 
 
 CO to O CO OV CT. OCTiCOTficoCM 
 M^OXOOmOWCONtON 
 
 MMM^tMMOCOrtCONCO 
 
 OOOrt! 
 OfflOt 
 
 WtOlOOffiMOCnOJtDlflCO 
 
 NC»OM'-000)(M-*tDT|t 
 C^D^O>COC7iCOOOOO>lO-*t<CO 
 
 N^cDNioaoHieowo 
 
 COi0 01COOOC000300iO'*t^ 
 
 MM 
 
 CD CD • • ■ CD 
 
 =3 3 : : : £8 8 8 : : : 
 
 o <d ■ i as c3 c3 • 
 
 « 0) tn m 03-h u ^i ^ * . 
 
 c e 88£2mm M+ ^^ 
 ^S^^ 3 ft & 1 22 2 2 
 
 •~"c-or^co^t , -5f<t^i— ittit^-^^< 
 
 COfOCNCOCOCOMCiJCOMCOM 
 
 o a> 
 
 ^a 
 
 §<! 
 
 5 
 
 &2c3oi.a.a 
 
 ct3 O „,T3 w o 
 O.D g CD O O 
 
 KNCOfiOtDI 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 67 
 
 also led to the conclusion that the seed of some of the species listed in table 8 
 are activated little or not at all by the heat treatment. Even when the seed 
 coats of such species were all but filed through, or when the testa was wholly 
 removed, as was done in some instances, and the seed subsequently placed 
 under germination conditions favorable to most seeds, the time of germination 
 was not hastened. 
 
 The seed of some of the chaparral species endured much higher tempera- 
 tures than did others. For example, the seed of only three species, namely, 
 chamise, common manzanita, and laurel sumac, showed any life when subjected 
 to temperatures between 280° and 300° F. The species whose heat-treated seed 
 gave much higher percentages of germination than the checks are chaparral 
 whitethorn, common manzanita, Parry manzanita, laurel sumac, bigpod cea- 
 nothus, sugarbush sumac, and western chokecherry. Although the testa of the 
 seed of some of these species, notably those of the two manzanitas and that of 
 the western chokecherry, is stony, thick, and dense, yet the interior of the seeds 
 of such structures reached the oven temperature in 3 to 4 minutes, as first 
 noted by Wright {117) and subsequently verified by the present writer. This 
 fact was determined by inserting thermocouples 0.006 inch in diameter 
 through closefitting holes made in the seed coat. 
 
 Table 8 shows that the grasses studied do not endure the higher temperature 
 ranges nearly so well as the chaparral species. No seed germination took place 
 among the five grass species when they were heated above 260° F. Moreover, 
 none showed appreciable increase in germination percentage under heat treat- 
 ment. This fact is best shown in those species whose germination check records 
 were relatively low, such as samples 1, 2, 6, 10, 11, and 12. Thus not only do 
 certain species of shrubs which often predominate on chaparral areas en- 
 dure high temperatures, but their germination capacity is sharply increased 
 thereby. The seed of other brush species, some of which constitute an impor- 
 tant part of the chaparral cover, are not stimulated to germination by the heat 
 treatments used. Although the seeds of grasses do not endure such high tem- 
 peratures as many associated shrubby forms, they do appear on new burns, 
 evidently from adjoining areas. The space occupied by established brush 
 seedlings soon becomes relatively great compared with that of the invading 
 grasses. The stimulated germination of the chaparral seed appears to account, 
 in the main, for the aggressive invasions of the chaparral species on burns. 
 
 Experimentation seemed to be justified by the extent to which the succession 
 of herbaceous vegetation on burns could be favored by the introduction of 
 seed of promising species. 
 
 RESEEDING OF BURNED AREAS 
 
 In 1926, reseeding experiments with cultivated forage plants were initiated 
 by the writer on burned chaparral and chamise lands in some of the northern 
 counties of the state. Since that time many such experiments have been under- 
 taken on burns. Some of these tests were made in the interior counties, where 
 the summers are hot and dry, and where the annual precipitation may vary 
 from 25 to 40 inches ; other such experiments were made in the cooler northern 
 coastal strip, where the annual rainfall ranges from 45 to more than 60 inches. 
 A large variety of burned brushland was included, particularly in the interior 
 
08 
 
 University of California — Experiment Station 
 
 counties ; in many instances ranch owners cooperated by furnishing areas for 
 reseeding. In addition, a large number of reseeding trials were made on public 
 lands and on railroad holdings. Sixty-two plots were established between 1926 
 and 1937, most of which were seeded to a mixture of two to five species. Most 
 of the seeding was done at the rate of 20 pounds per acre. The plots varied in 
 size from one-fortieth to one-half acre. 
 
 In these reseeding experiments many forage species were employed. In most 
 instances the seed was covered by raking or brushing it into the soil; but 
 where unusually hot fires had left the soil loose, or even "fluffy" by burning of 
 the humus, the seed was merely scattered on the ash-covered soil, with no 
 attempt at harrowing it in. The latter plan is popular along the north coast, 
 where some success has been attained in reseeding of burned brush ranges. 
 Both spring and fall seedings were attempted. In some instances the seed was 
 scattered early in the autumn before any rain had fallen; in other experi- 
 ments the sowing was delayed until after the first fall rains had thoroughly 
 wetted the soil and carried a portion of the soluble fraction of the ash into 
 the soil. 
 
 Reseeding in Interior Counties. — The interior counties included in the 
 reseeding trials were Butte, Glenn, Shasta, and Tehama. The species employed 
 were as follows : 
 
 Cultivated Grasses 
 Bulbous bluegrass 
 Canada bluegrass 
 Crested wheatgrass 
 Harding grass 
 Italian ryegrass 
 Meadow fescue 
 Orchard grass 
 Perennial ryegrass 
 Eedtop 
 Sheep fescue 
 Smilo 
 Timothy 
 
 Native and 
 Mediterranean Grasses 
 Blue wild-rye 
 California brome 
 Mouse barley 
 Purple needlegrass 
 Soft chess 
 
 Western needlegrass 
 Wild oat 
 
 Leguminous Species 
 Bur-clover 
 Spanish-clover 
 White sweet-clover 
 Yellow sweet-clover 
 
 All but two of the above-listed species are known to be growing without cul- 
 tivation in various parts of California, and some seem to be adapted to desert 
 habitats (79). 
 
 The results of these experiments were exceedingly disappointing. Trials 
 with most of the twelve cultivated grasses failed to produce practical forage 
 stands in the interior counties, regardless of season of seeding, covering of the 
 seed, the site quality, or the species used. Except for an occasional sparse stand 
 of some species, most of which endured for only two or three seasons, no suc- 
 cess was obtained. In this respect the results are similar to those recorded in 
 the Douglas-fir region of the Northwest (3) where, in order to retain the grass 
 cover, it is usually necessary to reseed every third year. 
 
 Generally, the few plants which survived the rigors of the first summer were 
 choked out in the second season by the abundant sprouts and seedlings of the 
 native brush cover, and by the weeds which temporarily invaded the burns. 
 Among the native perennial grasses, blue wild-rye and purple needlegrass 
 gave the greatest promise of success, although the stand was never dense. 
 
I Hi i, 68.1 J 
 
 Plant Succession on Burned Chaparral Lands 
 
 69 
 
 Among the annual grasses, wild oat and soft chess were the most successful. 
 Most of the other species listed above either failed to germinate, or the seed- 
 lings did not live through the first summer. Bulbous bluegrass, one of the 
 species not reported as growing without cultivation in California, was perhaps 
 the most persistent of any perennial grass on poor sites, but in no case was the 
 stand sufficiently dense to influence measurably the carrying capacity. Mouse 
 barley, though undesirable when mature because of the troublesome awns, 
 produced fairly dense stands on several plots the first season after seeding ; 
 but although viable seed was produced by this species, seldom were more than 
 a few plants found the second year. 
 
 TABLE 9 
 
 Height Growth and Density of Wild Oat Seeded on Burned Chaparral, 
 
 Shasta County, 1933 and 1934 
 
 Plot 
 
 Date 
 
 burned 
 
 Date seeded for first year's crop,* 
 and precipitation since burning 
 
 Density, 
 per cent 
 
 Average height at 
 maturity, inches 
 
 no. 
 
 First 
 year 
 
 Second 
 year 
 
 First 
 year 
 
 Second 
 year 
 
 31 
 32 
 33 
 34 
 
 Autumn 1932 
 Autumn 1932 
 Autumn 1932 
 Summer 1931 
 
 Autumn 1931 
 
 Summer 1931 
 
 Autumn 1931 
 
 Autumn 1932, after 3 inches of rain 
 
 Autumn 1932, after 1% inches of rain 
 
 Autumn 1932, before any rain 
 
 Autumn 1932, after about 35 inches of 
 
 40 
 35 
 
 27 
 
 8 
 12 
 
 4 
 10 
 
 5 
 2 
 3 
 
 
 
 2 
 
 
 
 5 
 
 51 
 54 
 47 
 
 11 
 
 14 
 
 16 
 
 13 
 
 13 
 12 
 9 
 
 35 
 
 Autumn 1932, after about 35 inches of 
 
 12 
 
 36 
 
 Autumn 1932, after about 35 inches of 
 
 
 37 
 
 Autumn 1932, after about 35 inches of 
 
 11 
 
 
 
 
 * The natural seeding from the first year's growth was augmented by scattering and raking in the same amount 
 of seed, in October, as was seeded on the plots the first year after burning. The resulting stand gave the second 
 year's crop. 
 
 Among the few species that gained even a temporary foothold, the height 
 growth and the density of the cover were superior on the plots burned during 
 the summer or fall of the year seeded, the densest cover and healthiest growth 
 being obtained on plots seeded after the rains had started. Wild oat, when 
 seeded in superior sites, was particularly conspicuous in its growth response 
 on areas heavily burned during the current season ; but in the second year the 
 vigor of growth and the density declined greatly, as shown by the measure- 
 ments presented in table 9. 
 
 Where the seeding of this species was delayed until the second year after 
 burning, the height growth was much lower and the stand thinner, indicating 
 that the soil nitrogen content, and perhaps the stimulating growth effect of 
 the ash, may have diminished. On plots 34, 35, 36, and 37, which were not 
 seeded until the autumn of the next year after burning, many chaparral 
 sprouts, seedlings, and weeds had reclaimed the area, so that light and mois- 
 ture were in unfavorable balance. Also the looseness characteristic of the 
 surface soil of hot, fresh burns was absent ; and the nitrate supply was meas- 
 urably lower than on recently burned areas, a factor discussed in connection 
 with the effect of fire on the soil. 
 
70 
 
 University of California — Experiment Station 
 
 The trials with leguminous species were generally no more encouraging than 
 were those with the cultivated grasses, as indicated in the following discus- 
 sion. Seeding with bur-clover and the sweetclovers resulted in consistent fail- 
 ures, whereas Spanish-clover produced a scattering of plants on some plots, 
 but failed entirely on others. 
 
 Beseeding in Coastal Counties. — The same species were used in the coastal 
 counties as were employed in the interior counties, and the same methods of 
 seeding were carried out. Work was done in Humboldt, Lake, Mendocino, and 
 Sonoma counties. 
 
 The results of these experiments were somewhat more successful than those 
 reported for the interior foothills, for temporary stands were often obtained 
 
 
 Fig. 28. — Gentle slope of heavy Douglas-fir stand cut and burned, followed 
 by seeding with 30 pounds of redtop per acre. A redtop cover of 60 per cent 
 density occupied the ground for the first two years after burning, both in the 
 protected enclosure and on the adjoining grazed area. In the third year annual 
 grasses and Aveeds replaced much of the redtop. In the fifth year after burning 
 and seeding, practically no redtop remained. The soil was deep and fertile. 
 Anderson Valley, Mendocino County, 1926. 
 
 in the better sites, particularly on plots close to the coast. Among the perennial 
 grasses, Italian ryegrass, perennial ryegrass, Harding grass, orchard grass, 
 redtop, and smilo gave somewhat consistently the best results (fig. 28). A 
 mixture of equal parts of seed of perennial ryegrass, orchard grass, redtop, 
 and smilo, sown at the rate of 20 pounds per acre, in sites of especially high 
 quality, proved the most satisfactory in the preliminary trials. Especially was 
 this the case in redwood cutover areas, where thickets of blue blossom were 
 first lopped and allowed to dry before being burned late in the fall, the area 
 being seeded after the coming of the first heavy autumn rain. In such sites the 
 seed was broadcast, with no attempt at soil treatment. The first year after 
 seeding, the yield on some plots was sometimes the equivalent of % to % ton 
 per acre. The forage was relatively lush and was much more highly relished 
 than that of the native annual grasses, or than that of the Mediterranean intro- 
 duced grasses. It was noted that the cultivated species retained their succu- 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 1 
 
 lence several weeks later in the season than did the annual grasses, hence there 
 was a tendency to overgraze them. By far the largest yield was produced the 
 first year. Indeed, the stand proved distinctly ephemeral, so that in the third 
 year most of the cover of cultivated grasses, regardless of its promising ap- 
 pearance the first year, was largely if not entirely replaced by the early- 
 maturing annuals, of which the inferior foxtail fescue usually predominated. 
 In a seeding mixture, redtop was usually the last to disappear. 
 
 Among the grasses which produced the heaviest yields on sites of excep- 
 tionally high quality, such as flats or gentle slopes of logged-off redwood forest, 
 Harding grass deserves special mention. Seeded in the autumn at the rate of 5 
 pounds to the acre, after the coming of the first rains, and without subsequent 
 
 Fig. 29. — Harding grass seeded on a burn of logged-over redwood on a 
 coastal range in Humboldt County. A good stand was procured for the first 
 two or three years after burning. 
 
 soil treatment, this species sometimes yielded heavily, despite its being pas- 
 tured closely (fig. 29). As in most other species, or seeding combinations, the 
 yield of Harding grass was heaviest the first year after seeding on fresh burns. 
 In the third season after seeding, the cover tended to decline measurably ; but 
 where grazing was conservative the stand gave promise of reasonable longev- 
 ity, in some instances possibly affording economic returns. Only sites of high 
 quality are suited to this species. The forage of Harding grass is palatable to 
 all kinds of livestock ; the other perennial grasses appearing in the list are 
 clearly of secondary value. 
 
 Because of the relatively greater promise of the perennial grasses in high- 
 quality sites, only a few plots were seeded to the annual grasses and to the 
 legumes. Wild oat and soft chess responded moderately well in localities too 
 poor for the growth of perennial grasses, as on the better chaparral burns 
 inland from the redwood belt. Except on deep soil, bur-clover and the other 
 legumes showed no response whatever. 
 
 The seeding trials of burned chaparral areas in the interior northern 
 
72 University of California — Experiment Station 
 
 counties, with the several species commonly employed in range revegetation, 
 indicated that reseeding of such areas is seldom economically feasible. Further 
 reseeding experiments, however, would be justified, on burns of good lands, as 
 more drought-enduring plants become available. The best possibilities of re- 
 seeding in the drier coastal areas appear at present to be with such native or 
 naturalized species as needlegrass, blue wild-rye, wild oat, and soft chess. In 
 the humid coastal areas, Harding grass and perennial ryegrass gave the best 
 promise of success. Even these species, however, frequently are soon crowded 
 out by the vigorously sprouting brush and by the numerous volunteer herbs. 
 
 EFFECT OF CHAPARRAL ASH ON SEED GERMINATION AND 
 GROWTH OF GRASSES 
 
 Reseeding of burned chaparral areas is frequently undertaken before the 
 autumn rains have dissipated the accumulated ash from the soil. Moreover, 
 reseeding experiments have been observed to give variable results on burned 
 areas where ash deposits are heavy. Recently burned areas frequently exhibit 
 spots barren of any form of vegetation for several years. On the other hand, 
 some chemical constituents of the ash appear to stimulate growth and develop- 
 ment of certain plants. In view of these confusing observations, it seemed 
 desirable to study the effects of ash upon germination and growth of some 
 grass species commonly selected for range reseeding. The study resolved itself 
 into three aspects : chemical analysis of the ash of selected chaparral species 
 to determine mineral composition of the ash before and after leaching; experi- 
 ments to note the effect of chaparral ash on grass-seed germination; and ex- 
 periments to determine the effect of ash on growth of grass from the seedling 
 stage to maturity. 
 
 Relatively little research has been done with respect to the effects of ash on 
 plant growth. Alway and Rost (2) reported that clover and timothy, when 
 seeded in the spring on an area burned the previous summer, produced abun- 
 dant forage the first year, but that this cover all but disappeared the second 
 season. Isaac and Hopkins (50) recorded low seedling survival on burned 
 forest lands, and the root systems of seedling plants were poorly developed. 
 Schmidt (93) reported that seed germination was lowered in direct propor- 
 tion to the amount of ash in the soil. Fabricius (25) applied thin layers of ash 
 of burned duff to forest soils. Of the seed of several species of trees planted on 
 the ashed soil, the germination of some appeared not to be affected, whereas 
 that of others was lowered. The survival of seedlings of all species was much 
 lowered by treatment with ash. 
 
 Harris and Pittman (31) conducted studies to determine the effect of alkali 
 on several common grasses. Potted soils were treated with varying amounts of 
 sodium chloride, sodium carbonate, and sodium sulfate. Each of these salts 
 proved toxic, and the degree of injury to the young plants decreased in the 
 order of the salts named. The species most seriously affected were Kentucky 
 bluegrass and timothy. Meadow fescue was only moderately affected, whereas 
 orchard grass and Italian ryegrass appeared indifferent to moderate amounts 
 of these salts. 
 
 Ash Content of Chaparral Species. — In the present study the chemical com- 
 position, both of individual chaparral species and of combinations of such 
 
tbul. 685] Plant Succession on Burned Chaparral Lands 
 
 73 
 
 species, was determined. Some samples were ashed in the field to simulate 
 burning practice, while other chaparral samples were burned under the hood 
 in the laboratory. The latter method resulted in somewhat more complete 
 ashing than that done in the field. The samples were collected in July from a 
 large number of bushes. The ash was derived from leaves, twigs, and stems in 
 such proportions as to be representative of entire bushes. 
 
 As shown in table 10, both the total ash and the silica-free ash are highest in 
 the leafy branch samples of blue oak, and lowest in wedgeleaf ceanothus. Blue 
 
 TABLE 10 
 
 Composition of Leafy Branches of Common Chaparral Species 
 
 (Expressed as per cent of ovendry weight) 
 
 Species 
 
 Total 
 ash 
 
 Silica 
 
 Silica-free 
 ash 
 
 Calcium 
 
 Phosphorus 
 
 Potassium 
 
 Wedgeleaf ceanothus 
 
 1.78 
 3.14 
 3 32 
 
 2.27 
 1.95 
 2.50 
 5 07 
 
 0.15 
 .39 
 .21 
 .34 
 .22 
 .31 
 
 0.24 
 
 1 63 
 
 2.75 
 3 12 
 2.93 
 1.73 
 
 2 20 
 4.83 
 
 0.716 
 1 141 
 1 082 
 0.707 
 614 
 630 
 1.200 
 
 0.060 
 .084 
 .083 
 .069 
 059 
 .071 
 
 155 
 
 0.255 
 0.490 
 
 
 0.709 
 
 
 0.335 
 
 
 0.475 
 
 
 0.683 
 
 
 1 209 
 
 
 
 TABLE 11 
 
 Composition and Alkalinity of the Ash of Common Chaparral Species when Burned 
 in the Open, to Simulate Natural Burning 
 
 Sample 
 
 Silica, 
 per cent 
 
 SiHca-free 
 
 ash, 
 per cent 
 
 Calcium, 
 per cent 
 
 Phos- 
 phorus, 
 per cent 
 
 Potas- 
 sium, 
 per cent 
 
 pHof 
 
 saturated 
 solution 
 
 Moles 
 of sodium 
 carbonate 
 per gram 
 
 of ash 
 
 
 6.1 
 28.3 
 4.9 
 9.6 
 6.6 
 6.9 
 11.8 
 
 90.1 
 64.1 
 92.6 
 86.4 
 89.1 
 89.4 
 85 6 
 
 34.9 
 27.7 
 30 2 
 24 4 
 25.3 
 23.6 
 25.0 
 
 2.68 
 1.58 
 1.98 
 3.10 
 2.27 
 1.99 
 2.97 
 
 8.99 
 9.50 
 17.24 
 15.10 
 19.08 
 17.25 
 10.92 
 
 12.4 
 11.1 
 12.4 
 11.4 
 12.4 
 12 4 
 11 6 
 
 0108 
 
 
 .0084 
 
 
 0114 
 
 
 .0090 
 
 
 .0103 
 
 
 0111 
 
 
 0124 
 
 
 
 oak is also highest in calcium, phosphorus, and potassium. Bigberry manzanita 
 and wedgeleaf ceanothus are the lowest in calcium and phosphorus ; the latter 
 species contains the lowest percentage of potassium of those studied. 
 
 Analysis of the ash of common chaparral species burned in the open to 
 simulate broadcast burning showed that chamise ash is much higher in silica 
 content than are any of the other species analyzed (table 11). The calcium 
 content is fairly constant in all the species analyzed, whereas the phosphorus 
 and potassium content vary more widely according to species. In all the ash 
 samples the calcium content is much higher than that of potassium, being more 
 than three times greater in some species. 
 
 Total alkalinity, for convenience, was expressed in moles of sodium carbon- 
 ate per gram of ash, but the values are seen to be so high that all the alkalinity 
 could not be due to carbonates alone. The carbonate values are greatest in the 
 
74 
 
 University of California — Experiment Station 
 
 madrone and blue oak, and lowest in chamise. With these high alkalinity data, 
 the recorded pH values, all in excess of 11, are not surprising. It seems likely 
 that the alkaline nature of the ash may be partly or largely responsible for the 
 germination and growth effects which are discussed later. 
 
 Leaching of Ash. — Knowledge of the solubility of the ash in water is of 
 importance in estimating the effect of rains in the leaching of ash on fresh 
 burns. Accordingly, known weights of ashed residues of sundry chaparral 
 species, collected in the first week of August, were placed over filter paper, 
 and slowly sprinkled with water in an amount equivalent to % inch of rain- 
 fall. The loss in weight, due to leaching, was then determined. The samples 
 were sprinkled a second time with the same amount of water and in the same 
 manner, after which the loss in weight was again noted. 
 
 TABLE 12 
 
 Solubility of the Ash of Common Chaparral Species when Sprinkled 
 with 1 Inch of Water 
 
 < 
 
 Species 
 
 Loss of salts 
 
 after first 
 
 J^-inch 
 
 sprinkling, 
 per cent 
 
 Loss of salts 
 after second 
 
 H-inch 
 
 sprinkling, 
 
 per cent 
 
 Total loss 
 
 from 
 
 sprinkling, 
 
 per cent 
 
 Loss of elements from the 
 two sprinklings 
 
 Calcium, 
 per cent 
 
 Phosphorus, 
 per cent 
 
 Potassium 
 per cent 
 
 Wedgeleaf ceanothus 
 
 11.75 
 11.68 
 25 00 
 16.13 
 26.70 
 25.70 
 12.07 
 
 18 48 
 
 1.10 
 
 2.22 
 2.19 
 3.19 
 2.50 
 1.31 
 1.27 
 
 1.97 
 
 12.85 
 13.90 
 27.19 
 19.32 
 29.20 
 27.01 
 13.34 
 
 2040 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 75.6 
 
 
 68.1 
 
 Pacific madrone 
 
 82.4 
 
 Dwarf interior live oak 
 
 65.1 
 
 Bigberry manzanita 
 
 80 
 
 Toyon 
 
 81.9 
 
 
 52.8 
 
 Average 
 
 72.8 
 
 
 
 Table 12 shows that the loss of soluble ash by leaching, through sprinkling, 
 was many times greater from the first half inch of sprinkling than from the 
 second half inch of sprinkled water. This would indicate that the first rain 
 after a fire would have much greater effect in removing soluble constituents of 
 the ash than would subsequent rains. The ash of different species, however, is 
 seen to respond to leaching somewhat differently. 
 
 As would be expected from the relative solubility constants of calcium, 
 phosphorus, and potassium in an alkaline medium, no appreciable loss took 
 place in the calcium and phosphorus contents of the leached ash, whereas 
 nearly three fourths of the potassium content was lost in the two slow sprin- 
 klings representing a total of 1 inch of water. The high solubility of potassium 
 in the ash suggests that this element may be a source of the stimulated plant 
 growth observed in greenhouse experiments and in the field. 
 
 Effect of Ash on Seed Germination and on Seedling Growth. — Effects of 
 chaparral and chamise ashes on seed germination and the seedling growth of 
 certain grasses were studied in the following manner : Seeds of each species 
 were shallowly planted in lightly compacted, well-moistened soil in terra cotta 
 pots, 10 inches in diameter. Before use, the pots were coated on the inside 
 with asphalt paint. Three ash treatments were given each species, as follows : 
 The seeds in some of the pots were covered to a depth of % inch with chamise 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 75 
 
 ash ; some of the pots were similarly treated with mixed ash consisting of equal 
 parts of chamise, wedgeleaf ceanothus, dwarf interior live oak, and blue oak ; 
 the seeds of the third group of pots were merely covered lightly with earth, to 
 serve as controls. 
 
 Half of the pots of each ash treatment contained red soil of the Aiken series, 
 whereas the rest of the pots contained yellow-brown loam of the Manzanita 
 series, both of which support extensive areas of chamise and chaparral. All 
 the pots were placed in terra cotta saucers. The pots with each soil, and each 
 ash treatment, were maintained at close to field capacity of moisture by keep- 
 ing the saucers filled with water, to avoid excessive leaching of the ash. The 
 rest of the pots were kept moist by sprinkling lightly from above, to simulate 
 rain. Germination was counted daily for 12 days, then the seedlings were 
 thinned to 6 plants per pot. Since there was little or no difference in the height 
 growth of the seedlings in the subirrigated and the sprinkled pots, or in the 
 two soil series, the measurements were averaged together. Measurements were 
 recorded at 7-day intervals for the first 4 weeks after the seeds were planted. 
 
 Figure 30 shows that orchard grass is the only species of the seven grasses 
 studied that gave considerably higher germination percentage in one of the 
 ashed soil groups, namely in chamise ash, than in the unashed soil. This 
 species, however, gave a correspondingly lower average germination percent- 
 age in the mixed-ash series. Germination of the seed of smooth brome was 
 nearly the same in the chamise-ashed soil as in the control series, but it was 
 lower in the mixed-ash soil than in the control set. The seed of redtop and 
 timothy was much lower in the ashed cultures ; and although crested wheat- 
 grass, Harding grass, and meadow fescue gave lower germination results in 
 the ashed soils, the decrease was less conspicuous than in redtop and timothy. 
 In all species the mixed ash had the most depressing effect on germination. 
 Thus the general effect of fair amounts of chaparral ash in the surface soil, in 
 this experiment, is seen to decrease the percentage of seed germination, and 
 in some species to delay sprouting of the seed. 
 
 The fact that seed germination in the chamise-ashed soil cultures is better 
 than in the mixed-ash appears to be correlated with another fact already 
 pointed out — namely, that chamise ash is much higher in silicon, and lower 
 in other minerals, per unit of weight of plant material, than the ash of any 
 of the other chaparral species analyzed. 
 
 In order to note the effect of ash dressings on the growth of the seedling 
 plants, the seed-germination experiment was continued 4 weeks after the first 
 leaf blades appeared. The results are presented in figure 31. 
 
 Seedling growth is seen to be less, as recorded by measurement of the length 
 of the leaf blades, in the soils treated with ash in six of the eight grass species 
 used. Crested wheatgrass and Harding grass exhibited slightly greater leaf 
 elongation in the chamise-ashed soil throughout the period of measurement. 
 Harding grass, however, made less growth in the pots treated with mixed ash, 
 and crested wheatgrass seedlings also grew more slowly in this ash until in 
 the fourth week, when they exceeded the growth of the control cultures. 
 
 Effect of Chamise Ash upon Growth to Maturity. — Whether the effect of 
 chamise ash would stimulate or actually retard the growth of grasses if the 
 measurements were carried to maturity of the plants, or whether the ash 
 
7G 
 
 University of California— Experiment Station 
 
 z 
 w 
 
 a 
 
 w 
 ex 
 
 . 10 
 2 
 
 
 1 ° 
 
 6 50 
 
 B 
 
 W 
 
 Q 
 W 
 
 8*> 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 :-- 
 
 --"" 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 „„- 
 
 — 
 
 
 
 
 
 i 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 „' 
 
 ,•' 
 
 
 
 
 
 
 
 
 
 
 • / 
 : / 
 • / 
 
 i 
 
 
 Ha 
 
 rdi 
 
 nq 
 
 gr 
 
 as< 
 
 
 
 
 
 
 
 7/ 
 
 Crested v, 
 
 I 
 
 r heatgra 
 
 ss 
 
 
 
 
 
 / 
 r 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 X 
 
 
 ... 
 
 — 
 
 — 
 
 — 
 
 *-- 
 
 
 
 
 
 ^• / ' 
 
 / 
 
 Tir 
 
 not 
 
 •fry 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 ■;> 
 
 
 it*- 1 
 
 
 
 " 
 
 '""" 
 
 ._. 
 
 ... 
 
 
 
 / 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 / 
 / 
 i 
 
 
 It 
 
 ali 
 
 an 
 
 ryegrass 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 ; Meadow 
 
 
 
 
 / 
 
 
 
 
 s 
 
 fescue 
 
 
 
 
 / 
 
 
 
 
 ' 
 
 
 
 
 
 
 
 
 / 
 
 
 
 / 
 
 * 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 A 
 
 
 ■' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t 
 
 
 
 '" 
 
 — 
 
 — 
 
 ~~ 
 
 
 
 
 
 1 
 
 
 /' 
 
 
 
 
 Drc 
 
 ha 
 
 rd 
 
 gr* 
 
 iSS 
 
 
 
 l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 " 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 1 
 
 
 
 
 
 
 
 
 
 
 Rg 
 
 dt 
 
 op 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Control 
 
 Chamise ash 
 
 Mixed ash 
 
 
 
 < 
 
 Smooth bromegrass 
 
 
 
 
 
 
 
 ^ 
 
 
 •"*" 
 
 
 
 
 
 
 
 
 
 
 *** 
 
 
 
 
 
 
 24-6 8 10 12 024 6 8 10 
 
 RATE OF GERMINATION, DAYS 
 
 Fig. 30. — Effect of ash on germination of seed of cultivated grasses. 
 
 would influence the length of the growing cycle of the grasses used, required 
 special study, and the use of large containers. Accordingly, a special series of 
 cultures were set up as follows : The grass seed was germinated in an incu- 
 bator ; and the resulting seedlings, selected for uniformity in size and apparent 
 vigor, were transplanted to galvanized iron garbage cans and filled with 
 moistened soil when the radicals of the seedlings were approximately % inch 
 
.Bul. G85] 
 
 Plant Succession on Burned Chaparral Lands 
 
 i i 
 
 long. The cans were 12 inches in diameter and 18 inches high. The soil used in 
 the experiments was sandy clay loam of the Manzanita series, collected under 
 a heavy stand of chamise in Mt. Diablo State Park. The soil was kept near field 
 capacity by sprinkling the surface with tap water. When the grass blades had 
 attained a height of y 2 to % inch, chamise-ash treatments were applied. 
 
 
 
 
 
 
 
 <£? 
 
 
 
 ^i^ 
 
 ■#>-" 
 
 '' 
 
 
 
 Harding 
 
 Grass 
 
 
 u I0 
 <D 
 
 .§ 5 
 
 c 
 
 8 ° 
 
 c 
 
 f 
 
 CT 5 
 
 '5 
 
 25 
 
 20 
 
 
 
 
 
 
 
 ^^& 
 
 
 
 ^ 
 
 ^"* . ■»■•■"" 
 
 
 
 Orchard Grass 
 
 
 
 
 
 
 
 
 
 
 
 -""" -*-.*i" 
 
 
 ^~°~ 
 
 
 Smooth 
 
 Brome 
 
 
 
 
 
 
 
 
 
 ..;/'' 
 
 
 
 s 
 
 
 ^.^ 
 
 
 
 Crested Wheatgrass 
 
 l l l 
 
 
 
 
 
 
 
 
 
 
 
 
 _^ 
 
 
 
 
 
 
 
 -^"* \ m 
 
 ^ ~~~~ 
 
 
 
 
 /^' 
 
 
 '" Time 
 
 >thy 
 
 
 
 Red 
 
 top 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 X 
 
 
 
 y 
 
 
 
 
 Italian 
 
 Ryegrass 
 
 
 LEGEND 
 
 Control 
 
 Mixed Ash 
 
 Chamise Ash 
 
 
 
 
 _/ 
 
 
 
 '•'^' 
 
 **'' 
 >^"" 
 
 
 
 '' 
 
 
 
 Meadov 
 
 / Fescue 
 
 
 12 3 4- 123 
 
 TIME IN WEEKS 
 
 Fig. 31. — Seedling growth under different ash treatments. 
 
 One complete set, including all the species studied, was grown without ash 
 treatment, to serve for control. One set including Harding grass, Kentucky 
 bluegrass, and perennial ryegrass was treated with ash by sifting it lightly 
 over the surface until the ash was % inch deep. One set of redtop and one of 
 soft chess, respectively, were treated with % inch of ash. Another set of redtop 
 and of soft chess received % inch of ash. A third set of redtop and of soft chess 
 
78 
 
 University of California — Experiment Station 
 
 received Vi inch of ash upon the surface of the soil, after which the ash of this 
 set was mixed with the upper 2 inches of soil. The plants in this last instance 
 were transplanted into the mixed soil. 
 
 Figures 32 and 33, which summarize these data, show that in the early 
 growth stages there was little difference in height in the five species studied, 
 regardless of ash treatment. It was not until between the third and fifth weeks 
 of growth that differences in size of the plants became apparent. After this 
 period the control plants of Harding grass, Kentucky bluegrass, redtop, and 
 perennial ryegrass grew faster and produced more leafage than did those 
 
 
 
 
 
 
 Treated specimen 
 
 .._ Kentucky Bluegrass (control) 
 
 Treated specimen 
 
 <•** 
 
 
 
 Pe 
 
 Tr 
 
 rennial r 
 eated sj 
 
 yegrass 
 secimen 
 
 (control 
 
 ' / 
 
 / 
 / 
 
 
 
 
 
 
 / 
 / 
 
 >•"""" 
 
 
 
 
 
 y^^'"' 
 
 
 
 
 
 / 
 / 
 / 
 / 
 / 
 
 
 / 
 / 
 / 
 
 / 
 
 
 .„.'" 
 
 '"' 
 
 
 
 / 
 
 / 
 
 / / 
 / / 
 
 ^'"' 
 
 / 
 
 / 
 
 
 
 
 
 A 
 
 f" / 
 
 
 / / 
 
 y' 
 
 
 
 „-"' 
 
 
 y .-"■■ 
 
 ,..~-""" 
 
 / 
 
 
 
 
 
 &:-'-" 
 
 
 
 
 
 
 
 53 68 83 93 
 
 DAYS AFTER PLANTING 
 
 Fig. 32. — Comparative growth of grasses when treated with 
 chamise ash. 
 
 treated with ^ inch of chamise ash. The only species which grew as well when 
 treated with % inch of ash, as under control conditions, was soft chess (fig. 
 33). Indeed, growth of this species was considerably stimulated by treatment 
 with Yg inch of ash. Mixing the ash with the soil stimulated the growth of soft 
 chess in the early growth stage, but this advantage was lost as the grass ap- 
 proached maturity. The field notes showed that flowering was affected ad- 
 versely by the ash treatment in all the species. The first evidence of flowering 
 was noted 93 days after planting, when 10 per cent of the plants in the control 
 cans were in the early flower stage, whereas in the treated cultures there were 
 merely a few flowers in the "boot" or sheath at that time, none being unfolded. 
 After 123 days, 75 per cent of the plants grown in the control cans were in 
 flower, whereas only 8 per cent of the treated plants displayed flowers. On 
 extensive chaparral burns, flowering was also observed to be slightly delayed. 
 During the first 6 weeks of growth the plants of the ash-treated cultures 
 were characterized by a dark bluish-green color of distinctive cast. However, 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 79 
 
 the leafage of the ash-treated soft chess plants began to show slight yellowing 
 after 59 days. This chlorotic condition continued to intensify until 123 days 
 after planting, when the herbage dried up just before flowering. In the con- 
 trols, inflorescence was much in evidence after 123 days, whereas the treated 
 plants showed but few flowers. Chlorosis of the leafage in the ash-treated red- 
 top series was noted 108 days after planting in the cultures treated with the 
 heaviest application of ash. This condition became correspondingly more pro- 
 
 
 
 
 
 
 
 
 
 ^.r,nt.roT 
 
 1 
 
 
 
 
 Treated with V A ' ash on soi 
 
 Tro»+.=><4 with V «^h or. <r> 
 
 1 
 I, 
 
 
 ^.^..-- 
 
 
 
 
 then mix 
 
 sd with ; 
 
 jpper 2 "o 
 
 : soil 
 
 ...-'"' 
 
 
 s,- 
 
 
 
 
 «■»■" 
 
 
 ^ 
 
 ~~^~~^ > 
 
 
 
 
 / 
 / 
 
 / / 
 
 
 
 
 
 
 
 // 
 
 
 
 
 
 
 
 
 
 ' 
 
 
 Soft 
 
 chess 
 
 
 
 ■z*''***'^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — _ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 // 
 
 ^-^~~ 
 
 ■^^' 
 
 
 
 
 
 ^r s''^' 
 
 / 
 
 
 
 
 
 
 _,_;.- 
 
 ^Z~&9 
 
 '1^^ 
 
 Rec 
 
 .top 
 
 
 
 
 
 
 
 
 
 
 53 68 83 93 
 
 DAYS AFTER PLANTING 
 
 Fig. 33. — Growth of soft chess and redtop treated with various 
 amounts of chamise ash. 
 
 nounced until the termination of the experiment, when approximately one 
 half of the ash-treated redtop plants had died. The untreated plants showed 
 no premature discoloration. 
 
 Composition of Treated Plants. — At approximate maturity the ash-treated 
 and the control samples of soft chess, redtop, and Harding grass were collected 
 for analysis, the sampling including plants of as nearly the same growth stage 
 as possible. The results are given in table 13. 
 
 None of the species studied showed appreciable differences as compared 
 
PQ «i 
 
 is- 
 
 ~ be 
 o £ 
 
 w > 
 « o 
 
 W ° 
 
 <« CD 
 
 CO o, 
 
 2 W 
 
 f^ o 
 51 
 
 
 «o o o> 
 
 O O! 1- 
 
 o o o 
 
 •o co co 
 
 CO CO © 
 
 3.76 
 
 3 53 
 3 45 
 
 481 
 445 
 437 
 
 •c im «c 
 |Q © «# 
 
 t- 
 
 CO 
 
 s 
 
 co 
 
 r- 
 
 to 
 
 eo 
 
 m 
 
 oo 
 
 
 CO 
 
 >* 
 
 IM — <M 
 
 S 6 * 
 
 goo 
 
 O cS ti 
 
 * % % 
 
 
 
 
 
 00 o 
 
 O O-l 
 
 
 -<j< »o 
 
 
 CO <M 
 
 
 o to 
 
 CO CO 
 
 o o 
 
 
 — i o 
 
 
 o o 
 
 CO "if 
 
 
 «o ■* 
 
 *H OS 
 
 
 »o >o 
 
 ■>»< eo 
 
 to i-H 
 
 o o 
 
 
 
 
 
 
 
 
 
 
 
 cS 
 
 
 s 
 
 £ x 
 
 
 
 rt a; 
 
 S c 
 
 $i 
 
 $ % 
 
 
 
 
 
 V ^ 
 
 & >. 
 
 
 
 
 
 ±; c 
 
 
 01 00 
 
 
 
 
 c3 33 
 
 
 0-, P_ 
 
 
 
 
 
 
 
 o 
 
 
 CO 
 
 
 -C 
 
 
 
 
 
 
 £ 
 
 
 -c 
 
 
 
 
 
 
 O g 
 
 
 t! -<= 
 
 
 O sJ 
 
 3^ 
 
 si 
 
 
 fe 
 
 S 
 
 F 
 
 o «- 
 co io 
 
 to en 
 to o 
 
 00 o 
 
 a .5 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 81 
 
 with the controls in either total ash or its constituents, as a result of the 
 various ash treatments. The crude protein content was also not affected. The 
 percentage of total ash was nearly the same in the treated and untreated plants 
 of soft chess and redtop. In Harding grass, however, the total ash was slightly 
 higher in the ash-treated plants because of the relatively high silica content. 
 Also the ash treatment did not appear to influence the proportion of calcium 
 and of phosphorus in the species studied, the ratio being maintained close to 1 ; 
 Gordon and Sampson (29) found the calcium : phosphorus ratio of approxi- 
 mately 1 to be characteristic of California grasses regardless of stage of 
 development. 
 
 Under field conditions the reaction of fire on the soil is not only reflected in 
 the liberation of ash, but also profoundly affects the activities of microorgan- 
 isms, some of which increase the fertility of the soil. 
 
 SOIL-NUTRITION AND PLANT-GROWTH STUDIES 
 
 BACKGROUND FOR THE STUDIES 
 
 It has long been contended that the burning of plant cover results in reduc- 
 tion of soil nutrients. Burning, it has been claimed, causes accelerated soil 
 erosion, the leaching of soluble minerals and nitrogen, and decreases the 
 organic matter of the soil. These deductions are supported by an extensive 
 literature. 
 
 More recent studies have indicated that the effect of burning on the soil is 
 not always detrimental to plant nutrition, but that it may, under certain con- 
 ditions, even enhance growth of some forms of vegetation (6). The liberated 
 ash, among other things, would appear to enrich the soil solution. Many soils, 
 however, are known to have strong fixing power for ash constituents ; hence 
 their liberation by fire may affect vegetation little or not at all (2) . Moreover, 
 if runoff and soil erosion proceed at a rapid rate, the top soil with its free ash 
 content may be largely washed away. 
 
 According to Shaw (95), Lowdermilk (64), and Larsen (60), erosion 
 usually carries off the finer soil particles first, which contain much of the 
 accessible plant nutrients; this exposes a lower layer, with its covering of the 
 coarser fractions. Sampson and Wehl (90) and later Forsling (26) found 
 that plants did not produce nearly so much growth in soil with the top layer 
 removed, as in soil with an undisturbed, well-formed A-horizon. Erosion on 
 heavily burned slopes is sometimes so serious as to remove much, if not all, of 
 the upper soil horizon, leaving the B-horizon, or subsoil, exposed. Thus Taylor 
 (106) concluded that destruction of soil fertility after repeated burning was 
 the result of erosion rather than the direct influence of heat. Sinclair and 
 Sampson (97) found that in the absence of the A-horizon soil the original 
 climax vegetation is exceedingly difficult to reestablish. 
 
 Heyward (37) found that the heating influence of fires in the longleaf pine 
 forests extends to only shallow soil depths. At a depth of % inch, temperatures 
 of 212° F were infrequent. At a depth of y^ inch the rise in temperature was 
 only a few degrees. Even these temperature rises lasted but a few minutes. The 
 extent to which burning may destroy the leaf mold was found, by Alway and 
 Rost (2), to vary widely in different soils. Rather wide variations in the total 
 volatile matter and in the nitrogen content of the leaf mold per unit area of 
 
82 University of California — Experiment Station 
 
 surface on burned lands was reported. Even exceedingly hot fires did not 
 consistently destroy the entire leaf mold, nor did they always greatly alter 
 the chemical composition of the soil solution. Fires with limited fuel supply 
 had little effect on the leaf -mold content of the soil, or on subsequent plant 
 growth. Moreover, Aldous (1), investigating the effects of burning on grass- 
 covered Kansas pastures, reported that the soil nitrogen content showed very 
 slight differences on burned and unburned plots during eight years of study. 
 Hensel (35), who also studied Kansas pastures, noted but slight change in 
 forage growth, but reported an increase in the botanical composition of the 
 herbaceous cover in early spring, followed by a decline in the number of 
 species with the advance of the season. 
 
 In a study of the southern longleaf pine belt, Hey ward and Barnette (39) 
 found that soils subjected to burning contained larger amounts of replaceable 
 calcium and were lower in hydrogen-ion concentration than unburned soils ; 
 but these differences were almost entirely confined to the upper 3 inches. They 
 concluded that the nitrogen and the replaceable calcium contents were higher 
 in newly burned areas ; that the pH values were slightly higher in the burned 
 plots ; and that no consistent differences were found in the organic matter. In 
 line with this study, Greene (30) concluded that the plant debris on the sur- 
 face is lost to the soil whether it is left to rot, or is burned. This worker 
 attributes the accumulation of the soil organic matter mainly to the decay of 
 plant roots. Greene also reported that soil nitrogen content and the organic 
 matter averaged somewhat higher on the burned plots early in the season. 
 Samples taken later in the season varied, being greater in either the burned or 
 unburned areas according to the season and the rapidity of plant growth on 
 the respective areas. Hey ward and Barnette (39) found that a period of eight 
 to twelve years is necessary to establish an approximate balance between the 
 accumulation and decomposition of the forest floor upon protection from 
 burning. After the balance is reached no increase in depth or weight of humus 
 on the forest floor occurs. Physical conditions under the humus layer follow- 
 ing protection from fire, they found, are favorable for plant growth, and the 
 humus layer on soils protected from fire appears in a healthy condition. 
 
 That certain plant species react more vigorously to the burned soil than do 
 others has been shown by several workers. Thompson (107) grew oats, pota- 
 toes, and sunflowers on forest soil, one area of which was cleared by the most 
 severe burning possible, whereas the check area was cleared merely by chop- 
 ping. The experiment was repeated annually for eight years. The oats and 
 potatoes yielded heavier on the unburned ground, whereas sunflowers yielded 
 heavier on the burned ground. Hesselman (36) likewise reported instances 
 where forest fires increased the rate of growth of young conifers. Where raw 
 humus had accumulated in thick layers fires were of some benefit by destroy- 
 ing part of the litter. 
 
 Although more or less indirect study has been made of the effect of burning 
 on the fertility of soils, the data are far from conclusive. Consistent differences 
 in the nutrition of burned and unburned soils are lacking, but in many in- 
 stances burning increases the nitrogen. Such effects as burning may have on 
 subsequent plant growth, however, appear to be confined essentially to the 
 surface soil layer. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 83 
 
 RESULTS OF CHEMICAL SOIL STUDIES 
 
 Soil Acidity, or pH. — Measurements of acidity, or pH of the soil, showed 
 no significant differences between that of the burned and unburned areas, 
 except in localized spots. The pH range was from 6.3 to 7.2 for the various 
 burned and unburned soils examined. In localized spots where much ash had 
 accumulated, the pH range was from 7.5 to 8.6. Except in small areas where 
 much ash had accumulated, the change in pH was apparently too slight meas- 
 urably to affect plant growth. 
 
 Types of Covers Whose Soils Were Analyzed for Nitrate Nitrogen. — Al- 
 though much nitrogen is volatilized when vegetation is burned, the total soil 
 nitrogen may not be greatly changed, since the decomposed roots, rather than 
 the top growth, apparently furnish the chief source of organic soil nitrogen 
 (30). The effect of burning on soil nitrate seemed worthy of study, as the 
 amount present might account partly, or even largely, for the identity, luxuri- 
 ance, and stability of the invading vegetation. This phase of the study was 
 initiated in Mendocino County in 1928, when the field and laboratory tech- 
 niques were developed, following which the study was extended elsewhere in 
 northern California. Samples were taken from the soil surface to 1 inch 
 depth, and at specified lower depths, to a maximum of 24 inches. 24 The study 
 included the sampling of four of the more extensive soil series, and of three 
 distinct plant covers, namely, the mixed chaparral, the pure chamise, and 
 areas dominated by interior live oak. 
 
 Nitrate Content of Mariposa Silt Loam in Mixed Chaparral Cover. — Here 
 the dominant vegetation consisted of greenleaf manzanita, California scrub 
 oak, and wedgeleaf ceanothus. The understory cover was composed of scat- 
 tered stands of foxtail fescue, nitgrass, rat-tail fescue, small-flowered lotus, 
 narrowleaf soap-plant, and Napa star thistle. The area sloped gently to the 
 south and east. The soil was fairly deep, and erosion had not exceeded the 
 normal rate. 
 
 The data obtained from the analyses of the soil samples, together with their 
 mean values, are presented in table 14 and are summarized in figure 34. This 
 figure compares, for three successive years, beginning in 1929, the trends of 
 nitrate nitrogen at different depths in the soils of burned and of similar 
 adjoining unburned areas. The Mariposa soil series here studied was sampled 
 in Mendocino County. Table 14 shows that the upper 1 inch of soil taken on 
 the burned plots contained an average for all such samples of 48.5 per cent 
 more nitrate the first year after burning than did the soil on the unburned 
 plots. On the other hand, the soil samples representing the lower soil depths 
 
 24 Paired burned and unburned plots were selected for sampling. All samples from each 
 specified depth were made up of composite soil collections, no one sample being: composed 
 of less than 7 random samples. Collections were obtained in various habitats. The earliest 
 samples were taken in the spring, when soil moisture and temperature favored initial vigor- 
 ous growth; a second sampling was done in early summer, when the major herbaceous 
 growth had been produced, but before the wilting point of the soil had been reached a foot 
 or so below the surface ; and a third series of samples was obtained in the autumn, when 
 the first 18 inches or so of the soil was dry, and the surface soil more or less baked. 
 
 Immediately after collecting, the samples were taken to the laboratory for analysis. Two 
 hundred grams of sieved soil were thoroughly mixed with 200 cubic centimeters of water for 
 5 minutes, then filtered through four layers of cheesecloth. Nitrate nitrogen in 15 cubic 
 centimeters of each of the turbid solutions was at once determined by the Devarda method. 
 
84 
 
 University of California — Experiment Station 
 
 TABLE 14 
 
 Nitrate Nitrogen Content of Soil in a Chaparral Area Burned in 1928, and of a 
 Similar Adjacent Unburned Area ; Mariposa Soil Series, Mendocino County 
 
 Date of sampling 
 
 Composite 
 
 sample 
 
 no. 
 
 Soil reaction, 
 pH 
 
 Burned Unburned 
 
 Nitrogen content, 
 p. p.m. NO.i 
 
 Burned Unburned 
 
 First year after burning (1929): sampling depth, 0-1 inch 
 
 March 4 
 
 1-2 
 
 7.3 
 
 7 2 
 
 39 4 
 
 20 5 
 
 92.2 
 
 April 16 
 
 3-4 
 
 7.6 
 
 7 4 
 
 46 8 
 
 18 
 
 160.0 
 
 May 19 
 
 5-6 
 
 7.4 
 
 7.4 
 
 29.3 
 
 25 1 
 
 16.7 
 
 June 11 
 
 7-8 
 
 7.3 
 
 7.2 
 
 33 7 
 
 29 4 
 
 14 6 
 
 July 13 
 
 9-10 
 
 7.3 
 
 7.5 
 
 39 
 
 31 .8 
 
 22 6 
 
 Sept. 10 
 
 11-12 
 
 7.3 
 
 7.3 
 
 21.1 
 
 16 2 
 
 30.2 
 
 
 
 7.4 
 
 7.3 
 
 84.9 
 
 28.5 
 
 48.5 
 
 
 First year after burning (1929) ; sampling depth, 1-6 inches 
 
 March 14 
 April 16 
 May 16 
 June 11 
 July 13 
 Sept. 10 
 
 Average. . 
 
 13-14 
 15-16 
 17-18 
 19-20 
 21-22 
 23-24 
 
 7.2 
 
 7.4 
 7.3 
 7.2 
 7.1 
 7.3 
 
 7.S 
 
 28.3 
 27.9 
 37 4 
 25.6 
 30.1 
 28.0 
 
 29.6 
 
 27.6 
 23.8 
 21.3 
 17.8 
 28.3 
 24.1 
 
 28.8 
 
 First year after burning (1929); sampling depth, 6-12 inches 
 
 
 25-26 
 27-28 
 29-30 
 31-32 
 33-34 
 35-36 
 
 7 
 
 7.2 
 7 3 
 
 7.4 
 7.3 
 7.2 
 
 7.2 
 
 7.3 
 
 7.2 
 7.0 
 7.3 
 
 7.2 
 7.2 
 
 7.2 
 
 34.3 
 29.6 
 22.3 
 29.8 
 19.5 
 23.4 
 
 26.5 
 
 28.7 
 27.2 
 31.6 
 23.8 
 24.7 
 18.0 
 
 25.7 
 
 19.5 
 
 April 16 
 
 8 8 
 
 May 16 
 
 -29.4 
 
 June 11 
 
 July 13 
 
 25.2 
 -21.1 
 
 Sept. 10 
 
 30.0 
 
 
 8.1 
 
 
 
 First year after burning (1929); sampling depth, 12-24 inches 
 
 March 14 
 May 1 6 
 July 13 
 Sept. 10 
 
 Average. 
 
 37-38 
 39-40 
 41-42 
 43-44 
 
 7 
 6.9 
 6.8 
 
 6.3 
 
 6.8 
 6.7 
 6.8 
 
 22.5 
 14.4 
 16.1 
 9.2 
 
 15.6 
 
 24 3 
 16.6 
 13 3 
 11.6 
 
 16.5 
 
 Second year after burning (1930); sampling depth, 0-1 inch 
 
 March 16 
 
 45-46 
 
 7 3 
 
 7.4 
 
 28.2 
 
 25 6 
 
 10 2 
 
 May 14 
 
 47-48 
 
 7 4 
 
 7.3 
 
 24 3 
 
 27 3 
 
 -11 
 
 Au-. 12 
 
 49-50 
 
 7.5 
 
 7.4 
 
 21 6 
 
 20.0 
 
 8.0 
 
 Average 
 
 
 7.4 
 
 7.4 
 
 24.7 
 
 24-3 
 
 1.6 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 TABLE 14— (Continued) 
 
 85 
 
 Date of sampling 
 
 Composite 
 
 sample 
 
 no. 
 
 Soil reaction, 
 pH 
 
 Burned 
 
 Unburned 
 
 Nitrogen content, 
 p. p.m. NO.-s 
 
 Burned Unburned 
 
 Per cen f 
 increase in 
 NOs wi,h 
 
 burnni 
 
 Second year after burning (1930); sampling depth, 1-12 inches 
 
 March 16 
 
 51-52 
 
 7.4 
 
 7.2 
 
 24.8 
 
 22 8 
 
 8.8 
 
 May 14 
 
 53-54 
 
 7.3 
 
 7.4 
 
 30 .6 
 
 27 6 
 
 10.9 
 
 Aug. 12 
 
 55-56 
 
 7.3 
 
 7.2 
 
 19 3 
 
 25 2 
 
 — 23 4 
 
 Average 
 
 
 7.3 
 
 7.3 
 
 24.9 
 
 25.2 
 
 -11 9 
 
 Second year alter burning (1930) ; sampling depth, 12-24 inches 
 
 March 16 
 
 57-58 
 59-60 
 61-62 
 
 7 
 
 6.8 
 . 6.8 
 
 6.9 
 
 6.9 
 6.8 
 
 6 8 
 
 6.8 
 
 15 9 
 12 
 10 2 
 
 12.7 
 
 18.3 
 13.0 
 12.1 
 
 lit. 5 
 
 -13 1 
 
 May 14 
 
 Aug. 12 
 
 Average 
 
 - 7.7 
 -15 .7 
 
 -12.4 
 
 Third year after burning (1931); sampling depth, 0-1 inch 
 
 March 16 
 
 63-64 
 65-66 
 67-68 
 
 7.4 
 7 3 
 7 5 
 
 74 
 
 7.5 
 7.5 
 7 4 
 
 7.6 
 
 25.8 
 
 26.2 
 22 
 
 84.7 
 
 23.8 
 27.6 
 25.7 
 
 25 7 
 
 8 4 
 
 May 15 
 
 Aug. 10 
 
 - 5.1 
 -14.4 
 
 A verage 
 
 - S.9 
 
 Third year after burning (1931); sampling depth, 1-12 inches 
 
 March 6 
 
 69-70 
 
 7 3 
 
 7 3 
 
 23 1 
 
 26.2 
 
 -11 .8 
 
 May 6 
 
 71-72 
 
 7.4 
 
 7 4 
 
 20 .2 
 
 19.3 
 
 4.7 
 
 Aug. 14 
 
 73-74 
 
 7.4 
 
 7.3 
 
 17.8 
 
 13.7 
 
 29.9 
 
 
 
 7.4 
 
 7 3 
 
 20.4 
 
 19.7 
 
 3.6 
 
 
 
 Third year after burning (1931); sampling depth, 12-24 inches 
 
 March 5 
 
 75-76 
 
 6.8 
 
 6.9 
 
 13.0 
 
 9.6 
 
 35.4 
 
 May 6 
 
 77-78 
 
 6.8 
 
 7.0 
 
 15 .7 
 
 13.0 
 
 20.8 
 
 Aug. 16 
 
 79-80 
 
 6.9 
 
 6.9 
 
 8.6 
 
 11.2 
 
 -23.2 
 
 Average 
 
 
 6.8 
 
 6.9 
 
 12.4 
 
 11.3 
 
 P.7 
 
 gave nearly the same nitrate values for both the burned and unburned plots. 
 In recording the values in figure 34 for the 1 to 12 inch samples, the nitrate 
 data for the first year after burning were averaged for the 1 to 6 inch samples 
 and for the 6 to 12 inch samples. 
 
 The nitrate values of samples taken from to 1 inch in depth, during the 
 second and third years after the fire, were not significantly different in any 
 burned and unburned areas. At depths of 1 to 12 and 12 to 24 inches, the 
 
86 
 
 University of California — Experiment Station 
 
 nitrate content was approximately the same on the burned and unburned plots 
 the first year after the fire. Likewise, the second and third years after burning 
 no effect on the nitrogen content was discernible. 
 
 Nitrate Content of the Hugo Clay Loam in Pure Chamise Cover. — Old, 
 dense chamise composed most of the cover. The nitrate nitrogen content of 
 field samples of this habitat, in Mendocino County, obtained from several 
 depths of the Hugo clay loam series, representing the first and second years 
 after burning, is presented in table 15, and summarized in figure 35. The small 
 amount of nitrate nitrogen in this soil series, despite the heavy growth of 
 chamise, is one of the most striking features revealed in this study. Despite 
 these low nitrate values, however, the trends paralleled those of the samples of 
 the Mariposa silt loam, as summarized in figure 34. Table 15 shows that the 
 
 3rd 
 YEAR 
 
 1 st 2 rid 3 rd 
 
 YEAR YEAR YEAR 
 
 1st 2nd 3rd 
 
 YEAR YEAR YEAR 
 12-24 INCHES 
 
 Fig. 34. — Nitrate nitrogen in field samples taken at three dif- 
 ferent soil depths in the Mariposa silt loam of the mixed chaparral 
 cover. Mendocino County. 
 
 samples taken during the first year after the fire, from the to 1 inch layer of 
 soil on the burned and unburned plots, averaged 6.2 and 3.0 parts per million 
 nitrate, respectively, or an increase of 106.7 per cent the first year after burn- 
 ing. Samples of the same depth, taken during the second year after burning, 
 revealed that the nitrate content had declined to an average of 3.9 parts per 
 million, whereas on the unburned plots there were 2.7 parts per million, or 
 an increase of only 44.4 per cent. Changes in the nitrate nitrogen content re- 
 sulting from burning appeared to be insignificant in the lower soil depths. 
 Thus the nitrate values of samples taken from the 6 to 12 inch and 12 to 24 inch 
 depths in the first and second years after burning are nearly the same in the 
 burned and unburned plots. In no soil sample did burning reflect any meas- 
 urable effect on the nitrate content in these lower depths studied. In fact, table 
 14 shows that only such slight differences as 2.2 and 1.8 parts per million of 
 nitrate were recorded in burned and unburned soils, respectively, when sam- 
 pled 1 to 6 inches deep and collected the first year after burning. 
 
 Nitrate Content of Aiken Clay Loam in Pure Chamise Cover. — Chamise 
 formed more than 90 per cent of the cover of this area in Shasta County. Here 
 and there scattered bushes of western mountain-mahogany, California buck- 
 
BUL. 685 j 
 
 Plant Succession on Burned Chaparral Lands 
 
 87 
 
 TABLE 15 
 
 Nitrate Nitrogen Content of Soil in a Chamise Area Burned in 1928, and of a Similar. 
 Adjacent Unburned Area ; Hugo Clay Loam Soil Series, Mendocino County 
 
 Date of sampling 
 
 Composite 
 
 sample 
 
 no. 
 
 Soil reaction, 
 pH 
 
 Burned 
 
 Unburned 
 
 Nitrogen content, 
 p. p.m. NO3 
 
 Burned Unburned 
 
 Per cent 
 increase in 
 NOs with 
 
 burning 
 
 
 First year after burning (1929); sampling depth 0-1 inch 
 
 
 
 March 7 
 
 81-82 
 83-84 
 85-86 
 87-88 
 
 6.4 
 6.4 
 6.5 
 6.4 
 
 64 
 
 6.3 
 6.5 
 6.5 
 6.4 
 
 6.4 
 
 12.1 
 6.3 
 4.1 
 2.2 
 
 6.2 
 
 4.6 
 3.0 
 2 2 
 2.3 
 
 SO 
 
 163.0 
 
 May 10 
 
 110.0 
 
 July 8 
 
 86.4 
 
 Sept. 11 
 
 - 4.3 
 
 
 106.7 
 
 
 
 First year after burning (1929); sampling depth 1-6 inches 
 
 March 7 . 
 May 10. 
 July 8. 
 Sept. 11. 
 
 Average. 
 
 91-92 
 93-94 
 95-96 
 
 6.3 
 6.5 
 6.4 
 6.4 
 
 6.3 
 6.4 
 6.3 
 6.2 
 
 3.6 
 2.2 
 1.1 
 1.8 
 
 2.9 
 1.4 
 2.0 
 1.0 
 
 24.1 
 57.1 
 -45.0 
 80.0 
 
 22.2 
 
 First year after burning (1929); sampling depth 6-12 inches 
 
 March 7. 
 May 10. 
 Sept. 11. 
 
 Average. 
 
 97-98 
 99-100 
 101-102 
 
 6.4 
 6.3 
 6.2 
 
 6.3 
 6.2 
 6.2 
 
 3.6 
 1.2 
 1.4 
 
 2.0 
 1.6 
 2.1 
 
 1.9 
 
 80.0 
 ■25.0 
 -33.3 
 
 10.5 
 
 First year after burning (1929) ; sampling depth 12-24 inches 
 
 March 7 
 
 103-104 
 105-106 
 
 6.2 
 6.5 
 
 6.4 
 
 6.3 
 6.4 
 
 6.4 
 
 2.1 
 
 3.6 
 
 2.9 
 
 3.2 
 3.8 
 
 S.5 
 
 -34.4 
 
 Sept. 11 
 
 - 5.3 
 
 
 —17.1 
 
 
 
 Second year after burning (1930); sampling depth 0-1 inch 
 
 March 10 . 
 May 8. 
 Sept. 18. 
 
 Average. . . 
 
 107-108 
 109-110 
 111-112 
 
 6.5 
 6.6 
 6.4 
 
 3.4 
 
 5.6 
 2.8 
 
 3 
 3.1 
 1.9 
 
 2.7 
 
 13.3 
 80.6 
 47.4 
 
 44-4 
 
 Second year after burning (1930); sampling depth 6-12 inches 
 
 March 10. 
 
 May 8. 
 Sept. 18. 
 
 Average. 
 
 113-114 
 115-116 
 117-118 
 
 6.3 
 6.3 
 6.5 
 
 6.3 
 6.4 
 6.3 
 
 3.0 
 
 2.7 
 2.1 
 
 3.3 
 
 2.4 
 1.9 
 
 - 9.1 
 12.5 
 10.5 
 
 4-0 
 
 Second year after burning (1930); sampling depth 12-24 inches 
 
 March 10 
 
 119-120 
 121-122 
 123-124 
 
 6.3 
 6.6 
 6.4 
 
 6.4 
 
 6.3 
 6.5 
 
 6.4 
 
 6.4 
 
 2.8 
 1.6 
 
 0.8 
 
 1.7 
 
 2.1 
 1.8 
 1.4 
 
 1.8 
 
 33.3 
 
 May 8 
 
 -11.1 
 
 Sept. 8 
 
 -42.9 
 
 Average 
 
 - 5.6 
 
88 
 
 University of California — Experiment Station 
 
 eye, and wedgeleaf ceanothus tended to break up the otherwise uniform ap- 
 pearance of the dominant shrub. The more common herbaceous plants were 
 foxtail fescue, nitgrass, downy chess, red larkspur, wild carrot, whispering 
 bells, and Fremont death camas. The area sloped to the west and had an aver- 
 age gradient of 12 per cent. The soil was shallow to moderately deep, and the 
 surface contained many rock fragments. Soil erosion appeared to be normal. 
 
 
 
 
 
 H Burned 
 
 
 
 ■ 
 
 
 
 
 ^^ Unhurried 
 
 1 
 
 Si 
 
 m\ 
 
 ■ 1 
 
 m -i 
 
 
 1st 
 
 YEAR 
 
 2nd 
 
 YEAR 
 
 1st 
 
 YEAR 
 
 2nd 
 
 YEAR 
 
 1st 2nd 
 
 YEAR YEAR 
 
 12-2-4 INCHES 
 
 
 O-l 
 
 INCH 
 
 6-12 
 
 INCHES 
 
 
 Fig. 35. — Nitrate nitrogen in field samples taken at three soil depthf 
 in the Hugo clay loam of the chamise cover. Mendocino County. 
 
 Burned 
 Unburned 
 
 2 3 
 
 YEARS AFTER BURNING 
 
 ig. 36. — Nitrate nitrogen in field samples at to 1 inch depth in 
 Aiken clay loam, chamise cover. Shasta County. 
 
 The analytical data of samples taken at the to 1 inch depth over a period of 
 five years following burning are summarized in figure 36. The nitrate content 
 shows the same general trends as those obtained for the soils of the Mariposa 
 and Hugo series, as summarized in figures 34 and 35. The sampling was done 
 in June of each year. It will be noted that the nitrate content of the burned 
 field samples was slightly more than twice that recorded in the unburned 
 samples the first year after the fire, but that this differential was nearly lost in 
 the second year after burning. In the following three years, the nitrate con- 
 tents of the soil samples representing the burned and unburned plots were not 
 significantly different one from the other. This evidence indicates, as in the 
 previously discussed soil series, that the effects of burning on nitrification of 
 the soil take place essentially only the first year after the fire. The nitrate con- 
 tent of the Aiken soil series is seen to be low and corresponds closely in this 
 respect to that of the Hugo clay loam soil series where chamise predominated. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 89 
 
 Nitrate Content of Los Osos Clay Loam in Interior Live Oak Cover. — This 
 flat area, located in Shasta County, had moderately deep to deep soil, and 
 supported a heavy stand of interior live oak, with secondary woody plants of 
 Pacific madrone, hoary manzanita, whiteleaf manzanita, and coffeeberry. The 
 herbaceous vegetation consisted chiefly of foxtail fescue, ripgut grass, sheep 
 sorrel, coyote tobacco, field suncup, Napa star thistle, and Spanish-clover. Soil 
 erosion had apparently not exceeded the normal rate. 
 
 Since previous studies had revealed that burning did not appreciably affect 
 the nitrate content at lower soil levels, the sampling in this series was confined 
 to to 2 inches in depth. The samples were first collected in 1933, which was the 
 first year after the fire, and were continued for three successive years. The 
 averaged data gave the following results : first year after burning, 39 parts 
 per million, as compared with 24 parts on the control plot ; second year after 
 burning, 21 parts per million, as compared with 22 parts on the control area ; 
 third year after burning, 19 parts per million, as compared with 22 parts in the 
 control samples. The levels of the nitrate content of this relatively productive 
 soil series are seen to be fairly high for chaparral soils. Also on this burn the 
 increase in nitrate nitrogen was significantly high only the first year after 
 burning. After the first year the nitrogen content on the burned plot leveled 
 off to nearly the same values as on the control area. 
 
 Nitrate Content as Affected by Exposed but Not Burned Chaparral Soils. — 
 It seemed important to obtain some indication whether the increase in soil 
 nitrate is associated with the heat created by burning heavy brush stands, or 
 whether significant nitrification would result from mere exposure of soil 
 to the full play of the sun. Accordingly, in the autumn of 1932, a heavy 
 chamise growth, in Mendocino County, on an area 80 feet square, of the Aiken 
 soil series, was cut at the surface of the ground and removed from the plot, 
 leaving the soil fully exposed to the sun. The soil sampling, taken at to 2 
 inches in depth, was done in spring, summer, and autumn for three successive 
 years after removal of the top growth. 
 
 The nitrate content was found to be affected little or not at all by the brush- 
 removal and soil-exposure treatment. The first year after removal of the brush 
 there was an average of 7.9 parts per million of nitrate in the spring, summer, 
 and autumn samples, as compared with 6.7 parts in the samples of the control, 
 or un chopped adjoining chamise plot ; and in the following two years the data 
 were even more nearly of the same values. It was significant too, perhaps, that 
 only a few chaparral seedlings appeared on the chopped plots, despite the fact 
 that surface soil temperatures reached 160° F on several occasions during the 
 hottest part of the summer. 
 
 Comparative Nitrate Content of Different Covers. — The nitrate nitrogen 
 content in the shallow soil samples which supported mixed chaparral was con- 
 sistently higher after burning than in similar soil samples which supported 
 the pure chamise cover. Slightly higher than in the mixed chaparral soils, 
 however, was the nitrate content of the soils which produced a heavy growth 
 of interior live oak. Nevertheless, all soil samples from these plant covers 
 showed perceptible increases in nitrate nitrogen in the surface layer the first 
 year after burning. In the second and subsequent years after burning, the 
 nitrate content declined so sharply as closely to approach the levels in un- 
 
90 University of California — Experiment Station 
 
 burned soils. Some evidence was obtained to indicate that the increase in 
 nitrate content following burning may be associated with soil productivity, 
 and with the heat of the fire. The deeper, more productive soils seemed to build 
 up considerable nitrogen after burning; but even such soils showed little 
 response when sampled on areas which had been burned so heavily as to have 
 been left more or less sterilized. 
 
 By way of explaining the increased nitrate content of the surface soil after 
 a fire : it is possible that the heat of the fire destroys the microflora and micro- 
 fauna of the surface soil layer. Later, unusually active colonization of nitro- 
 gen-fixing and other bacteria may be favored until organisms, which feed 
 extensively or exclusively upon the large numbers of bacteria, also come up 
 
 Fig. 37. — Hill lotus, showing proportionate sizes of plants: A, grown on unburned 
 chamise area; B, on chamise land burned two years previously; C, on chamise area burned 
 the previous summer. 
 
 from lower soil depths, and multiply at an unusually rapid rate. This hypothe- 
 sis has been suggested by Hesselman (36). 
 
 The ecological significance of the increased soil nitrate resulting from burn- 
 ing would obviously be most conspicuously reflected in the growth response of 
 those plants which feed largely in the surface soil layer. For example, the 
 numerous shallow-rooted annual herbs, which usually appear in greatest num- 
 bers the first year after a fire, are most favored by the increased nitrate of 
 freshly burned areas. The characteristic robust growth of these plants in the 
 first year after burning, and the striking sharp decline in their size and num- 
 bers during subsequent years, parallel significantly the sharp increase and 
 decline in nitrate which has been described. 
 
 CHEMICAL COMPOSITION OF VEGETATION ON BURNED AND 
 
 UNBURNED AREAS 
 
 Domestic foraging animals, as well as native herbivores, have been observed 
 to feed upon newly-burned chaparral and chamise areas in preference to 
 adjoining unburned lands. Whether this behavior may be attributed to the 
 
Jim, 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 91 
 
 fact that the feed is more readily available because of the reduced height of 
 the brush, whether the change in the species composition is a factor of impor- 
 tance, or whether the claim is valid that "the forage is sweeter and more nutri- 
 tious," is problematical. 
 
 Plant Succulence. — A rather consistent difference in the vegetation pro- 
 duced on newly burned areas, as compared with that of unburned lands, is the 
 more rapid growth, greater volume of individual specimens, somewhat higher 
 
 Fig. 
 
 38. — Shoots of blue oak: A, from one-year-old burn; B, from two-year-old burn; and 
 C, from an adjoining unburned area. (About % natural size.) 
 
 moisture content of the herbaceous plants, and more numerous production of 
 vigorous shoots of shrubby plants for the first year or two after burning (figs. 
 37, 38, and 39). The decreased competition for soil moisture and the stimula- 
 tion produced by the addition of nutrient salts in the ash probably account 
 for the more rapid and luxuriant growth and the higher moisture content of 
 
92 
 
 University of California — Experiment Station 
 
 the plants on burned areas. Table 16 shows the moisture relations of some 
 plants common to burned and unburned chaparral lands. 
 
 In the early growth stage only small differences in the percentages of mois- 
 ture in favor of the plants collected on the burns were recorded in the shrubs 
 and herbs. A wider difference in the percentage of moisture was noted, how- 
 ever, with the advance of the season, followed by a decrease and a tendency 
 toward equalization in moisture percentage towards midsummer, when the 
 vegetation approaches maturit}^. This difference in succulence might partly 
 account for the forage choice shown by the animals on the burned areas in 
 March and April, when the succulence of the vegetation has declined meas- 
 urably on the unburned areas. The character of the herbaceous vegetation, 
 especially with respect to the extent and deptli of the root systems in propor- 
 
 Fig. 39. — Shoots of yerba santa: A, from an unburned area; B, from an adjoining 
 one-year-old burn. (About % natural size.) 
 
 tion to aerial growth, may account for the differences in succulence in some 
 plants on burned and unburned areas. The difficulty of finding plants of 
 exactly the same developmental stage on the burned and unburned plots must 
 also be considered in explaining these differences in percentage of moisture. 
 Chemical Composition. — The species sampled for chemical study were those 
 most commonly found on chaparral and chamise lands. The samples of the 
 species employed were taken at random, the collections being made at the par- 
 ticular growth stage most in evidence at the time. The somewhat delayed 
 maturity of the vegetation on the burned areas made it difficult to procure 
 samples of exactly the same growth stage as on the unburned plots. There 
 appeared to be an average lag of about 1 week in plant development of most 
 herbaceous species worked with on the burned areas studied. Perhaps a similar 
 lag in development occurred in the sprouting shrubs; but no concrete differ- 
 ence could be recognized, since the young sprouts and seedlings on the burned 
 areas do not flower the first year after a fire. Most samples of the shrubs and 
 trees consisted of leafy stems or the foliage, which was stripped from the stems 
 
Bul. G85] 
 
 Plant Succession on Burned Chaparral Lands 
 
 93 
 
 ii 
 
 * i-, 
 3& 
 
 Q6 
 
 CO 
 
 CO 
 
 00 
 
 H 
 
 O 
 
 cc 
 
 rH 
 
 CO 
 
 ee 
 
 o 
 
 00 
 
 r-l 
 
 Tf 
 
 oi 
 
 © 
 
 >o 
 
 CO 
 
 
 ,_ 
 
 
 g 
 
 
 s 
 
 
 
 
 
 
 
 on 
 
 
 C3a 
 
 
 
 «5 
 
 
 -r 
 
 «© 
 
 
 ^ 
 
 to 
 
 ■o 
 
 •^ 
 
 -f 
 
 5C 
 
 
 CN 
 
 CO 
 
 CO 
 
 CO 
 
 
 t~ 
 
 • O 
 
 
 
 CI 
 
 <# 
 
 cr- 
 
 CJ 
 
 
 CO 
 
 CO 
 
 00 
 
 co 
 
 OS 
 
 CM 
 
 _i 
 
 r_ 
 
 
 OJ 
 
 r^ 
 
 ,_( 
 
 
 cr 
 
 on 
 
 
 s§ 
 
 eo 
 
 
 Its 
 
 rH 
 
 1, 
 
 o 
 
 co 
 
 «o 
 
 -*■ 
 
 «o 
 
 SO 
 
 -J? 
 
 ■o 
 
 iO 
 
 so 
 
 >c 
 
 
 1- 
 
 
 CO 
 
 
 -<t< 
 
 t; T) CD <p o £ 
 
 8 1 1 
 
 CD > 
 
 > cd 
 
 -2 -0 
 
 a — c S ^ i? i? ~ 
 
 © © o o o 
 > > > > > 
 
 cJ oj d cj cj 
 
 
 
 
 TJ 
 
 
 
 
 
 "C 
 
 -r 
 
 -0 
 
 ' _ 
 
 >. 
 
 >. 
 
 >. 
 
 ,4 
 
 3 
 
 - 
 
 3 
 
 _0> 
 
 
 
 
 
 CD 
 
 CD 
 
 cu 
 
 
 > 
 
 > 
 
 
 
 a 
 
 Rl 
 
 Kt 
 
 3 
 
 ££X 
 
 J 
 
 § s 
 
 CO CO 2 
 
 " O CJ o 
 
 "c3 13 13 -Q 
 
 W O O _ 
 
 £ -S 
 
 £ g 
 
 _0 CD 
 
 cd t! ® «> ^ <d 
 J £ B W £ W 
 
 ■ • -33333 
 
 M_ *4_, <*_i ►>**-,«*_, <*_, >**_,«*_*«-, >> ?* P* P* ►> r* 
 
 *< S 1-5 *-S "< <5 
 
 <B CD -r-i 
 
 a a c3 S 2 a cj 
 
 <! S 
 
 ^ it H fc. 
 
 e e 3 « 
 X! -Q -O -O 
 
 o o o o 
 
 
 
 
 a 
 
 E 
 ■j 
 a 
 
 DC 
 
 a 
 E 
 
 a 
 
 a. 
 
 (0 
 
 3 
 
 a 
 
 en 
 
 ^ -M M -* 
 c3 c3 c3 & 
 O O O O 
 
 3 3 3 3 
 
 £ £ a> a> 
 
 §.££££ 
 
 3 xi xi ji js x\ * * a .-a a 
 
 t^ 
 
 oc. 
 
 CO 
 
 Ol 
 
 •H 
 
 CO 
 
 _l 
 
 ro 
 
 _* 
 
 to 
 
 _ 
 
 
 
 
 CO 
 
 
 iO 
 
 <M 
 
 CO 
 
 
 t- 
 
 oo 
 
 CO 
 
 00 
 
 
 _ 
 
 ,f 
 
 
 on 
 
 
 oo 
 
 CO 
 
 CO 
 
 
 >o 
 
 (M 
 
 to -^* 
 
 cd 3 
 
 t; "« "£ t- t3 "£ 
 
 en co co on oo co 
 
 S £ S £ a s 
 
 CO CD CD CD G> CIS 
 
 CO co oo CO CO CO 
 
 <: s -, << g 
 
 ^3 pS 
 
 © CD CD "® "« *® 
 > > > J2 JD ^2 
 
 J2 ,2 ^ M bO bC 
 
 C o o 3 3 3 
 
04 
 
 University of California — Experiment Station 
 
 Q 
 
 l-l XJ 
 
 XJ 
 
 § s 
 
 
 paujnqufi 
 
 paiung 
 
 XTep 
 
 So* 
 ^a 
 
 paun.iqnfi 
 
 pau.mg 
 
 
 paUJIiqUf} 
 
 paujng 
 
 o 
 a 
 
 CO 
 
 o 
 Xi 
 Ph 
 
 paumqiiQ 
 
 paiiing 
 
 a 
 o 
 
 paiunqufj 
 
 pau.mg 
 
 £ 
 
 GO 
 
 03 
 
 O 
 
 PL, 
 
 paujnqufj 
 
 paujng 
 
 0) 
 
 c3 to 
 CO 
 
 paiunqiij} 
 
 pau\mg 
 
 0> 
 
 rd 
 
 5* 
 
 o 
 
 M 
 
 X 
 
 3 
 
 t, 
 a 
 3 
 
 Date of 
 sampling 
 
 U a) 
 
 
 m 
 
 
 OCBOINIDOWOO'* 
 
 
 CO ^~ 
 
 *00«NCtC!)0;nN 
 
 
 co i- 
 
 •^rf^m^McoNW 
 
 
 
 
 
 
 
 N n 0O 
 
 ffiHN!OMK5!DM**NN» 
 
 CM « ■* CM us es 
 
 ■Tf< CO »C -^ 
 
 W N M 
 
 (M «- 
 
 CO CM t-h CM CM CM CN 
 
 CM >- 
 
 CO CM 
 
 O CM r- 
 
 
 ,-H r- 
 
 co r- 
 
 
 C35 ■"*! CO CM 
 
 
 r-- os os co -d 
 
 
 ioew«iN(»!OininN»c 
 
 " 
 
 
 
 
 
 t^ CO © CO CO © "- 
 
 CO t~- ^ 
 
 
 05 CM 
 
 «5 S M < 
 
 CO©t^»0»000©eO»« 
 
 
 
 
 '- 1 
 
 
 
 
 
 
 
 
 00 — 
 
 OS — 
 
 
 
 O O ■* !D 
 
 CM r~ © O CM i- 
 
 
 
 CO CM 
 
 ■* * IO «J 
 
 co co c- 
 
 lO T- 
 
 CM CO C<" 
 
 OS 
 
 Tt< OS CN 
 
 n •* a es oo ■« 
 
 »o OC 
 
 © OS 
 
 ■<»<»ri©cDCOOOCOOO«Oir- 
 
 >o c«- 
 
 
 •^ -«t 
 
 lO ■*» 
 
 CO CN 
 
 CO ^ 
 
 •* 
 
 IO CN 
 
 CO ■«< 
 
 
 OS N ^ 
 
 
 (M 
 
 
 co 
 
 •* O) u- 
 
 
 co oc 
 
 cr 
 
 c 
 
 K 
 
 r^ 
 
 CO 
 
 C^ 
 
 
 © 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 CC 
 
 
 ir 
 
 ec 
 
 
 5 
 
 cr 
 
 -r 
 
 CM 
 
 
 
 
 
 
 
 
 i^ 
 
 
 
 
 
 
 
 
 
 
 
 e 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 o 
 
 CO t~- 
 
 CT 
 
 
 CO CM 
 
 Ol 
 
 h- 
 
 CT 
 
 c- 
 
 c 
 
 C 
 
 <M 
 
 
 
 
 
 
 
 e« 
 
 
 ec 
 
 
 
 eo cr 
 
 
 •c 
 
 «5 »C 
 
 
 IC 
 
 c 
 
 cc 
 
 
 
 o> 
 
 o 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 »o oc 
 
 li- 
 
 or 
 
 
 ir 
 
 iC 
 
 
 >d 
 
 c 
 
 
 
 ce 
 
 
 CO »c 
 
 ^r 
 
 
 
 
 
 
 CO 
 
 o c 
 
 c 
 
 o 
 
 o o 
 
 cr 
 
 o 
 
 O 
 
 O 
 
 o 
 
 O 
 
 
 
 
 — 
 
 r-i oc 
 
 
 :r 
 
 
 
 r^ 
 
 r-- 
 
 § 
 
 t~- oc 
 
 
 ec 
 
 t- oo 
 
 CO 
 
 r ^- 
 
 c^ 
 
 oc 
 
 
 
 o o 
 
 o 
 
 o 
 
 
 ° 
 
 "~ 
 
 
 •^ 
 
 1-1 
 
 CM 
 
 rt 
 
 
 s 
 
 1 - 
 
 O O- 
 
 s 
 
 eo 
 
 00 
 
 CO 
 
 e 
 
 -r 
 
 co 
 
 r- a~ 
 
 
 ■^1 t^ 
 
 
 
 
 
 
 
 o o 
 
 c 
 
 c 
 
 
 c 
 
 
 •^ 
 
 
 " 
 
 C) 
 
 " 
 
 
 
 
 
 
 
 
 
 
 1^ 
 
 CO 
 
 t- r- 
 
 
 t- 
 
 OS CM 
 
 K 
 
 CT 
 
 CO 
 
 I.-C 
 
 oo 
 
 D 
 
 
 co •* 
 
 CM 
 
 CM 
 
 CO CO 
 
 •*r 
 
 ■f 
 
 OC 
 
 CO 
 
 CO 
 
 
 lO 
 
 oo cm 
 
 
 
 
 
 or 
 
 
 1^ 
 
 
 
 CO 
 
 cr 
 
 cr; 
 
 
 
 oc 
 
 
 
 01 
 
 
 
 CM ■>*■ 
 
 o 
 
 CN 
 
 CO CO 
 
 •»f 
 
 ■* 
 
 CO 
 
 CO 
 
 CC 
 
 
 >o 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 »■ 
 
 
 a 
 
 
 F 
 
 £ F 
 
 Fl 
 
 E 
 
 F 
 
 F 
 
 
 ' fi 
 
 
 
 a 
 
 
 
 
 
 
 
 
 IK 
 
 
 
 
 
 
 
 
 
 
 
 
 ll 
 
 
 
 
 '/ 
 
 
 
 
 
 
 
 6 
 
 > 
 
 J >» > 
 
 1 > 
 
 > 
 
 . > 
 
 , > 
 
 g 
 
 F 
 
 >-. 
 
 
 
 
 
 
 
 
 
 ca CC 
 
 a 
 
 * 
 
 C3 C! 
 
 
 r 
 
 s 
 
 3 
 
 e* 
 
 a 
 
 c3 
 
 
 
 fl 
 
 
 a 
 
 0. 
 
 (1 
 
 a 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■+J 
 
 -*i -^ 
 
 
 
 
 
 
 
 
 3 c 
 
 c 
 
 c 
 
 c: c 
 
 c 
 
 
 c 
 
 r 
 
 r 
 
 a 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 b 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t 
 
 « 
 
 u 
 
 (H 
 
 3 3 
 
 o c 
 
 
 s 
 
 3 3 
 
 - 
 
 = 
 
 3 
 
 3 
 
 a 
 
 
 3 
 
 3 
 
 o 
 
 ,_, ,_, 
 
 ^_ 
 
 e> 
 
 CM CM 
 
 <M 
 
 Tl 
 
 -r 
 
 cr 
 
 CO 
 
 CO 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 OS CT 
 
 ~ 
 
 cr 
 
 OS 03 
 
 — 
 
 C~ 
 
 — 
 
 c 
 
 c- 
 
 ■cr. 
 
 o 
 
 t>-" r- 
 
 l~ 
 
 CO 
 
 ■* O: 
 
 rt 
 
 -r 
 
 ^ 
 
 CC 
 
 CO 
 
 co 
 
 _: 
 
 
 
 
 .3 
 
 
 " 
 
 
 
 
 
 
 a c 
 o> co 
 
 CO CO 
 
 c 
 
 X 
 
 > 
 o 
 
 c3 c3 
 
 6 
 
 < 
 
 > 
 O 
 
 y 
 
 x5 
 
 > 
 
 OS 
 
 > 
 
 > 
 
 > 
 
 +5 
 
 O 
 
 
 
 
 
 
 
 >o 
 
 >c 
 
 US 
 
 «c 
 
 «o 
 
 
 
 
 
 
 & 
 
 
 
 
 -" 
 
 
 OS OS 
 
 c-- 
 
 c-- 
 
 Ol O 
 
 >_. 
 
 o. 
 
 c. 
 
 C3J 
 
 c. 
 
 '-•- 
 
 
 
 
 
 
 
 
 c3 
 
 OS 
 
 g. 
 
 cr 
 
 c3 
 
 
 
 
 
 
 
 
 
 
 
 
 o o 
 
 o 
 
 o 
 
 o o 
 
 o 
 
 o 
 
 
 
 
 
 
 s s 
 
 e 
 
 fl 
 
 
 
 
 o 
 
 O 
 
 O 
 
 C 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 o c 
 
 c 
 
 c 
 
 o c 
 
 r 
 
 
 
 
 
 
 
 X X 
 
 X 
 
 X 
 
 -3 X 
 
 X 
 
 T? 
 
 p 
 
 (^ 
 
 P 
 
 
 
 3 s 
 
 
 CI 
 
 3 3 
 
 c 
 
 3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 s 
 
 a 
 
 a 
 
 s 
 
 a 
 
 s 
 
 !^ 
 
 o 
 
 o 
 
 u 
 
 U 
 
 o 
 
 
 *H 
 
 00 
 
 o 
 
 2 
 
 cc. 
 
 CM 
 
 CO 
 
 O 
 
 CM 
 
 co 
 
 3 
 
 <M 
 
 lO »o (^ 
 
 ^-2 3 
 
 Jr > * 
 ^ S 
 
 JJhi 
 
 s a s 
 
 oc 
 
 
 •"CC 
 
 t>. 
 
 
 o- 
 
 o -< 
 
 
 
 ^ 
 
 
 
 
 CO CO 
 
 
 CM 
 
 
 
 ~ 
 
 
 OS T«« 
 
 CO CM 
 
 1^. 
 
 1 - 
 
 
 CN 
 
 — 
 
 
 c^ 
 
 CM CM 
 
 
 CO © 
 
 oc 
 
 OS © 
 
 
 
 ' - 
 
 
 OS CM 
 
 
 
 e 
 
 OO CM 
 
 
 
 rt 
 
 tr. 
 
 su 
 
 c 
 
 © CM 
 
 
 
 ^ 
 
 
 
 
 
 CO CO 
 
 
 
 I-H 
 
 c 
 
 
 © 
 
 
 co oo 
 
 
 
 
 CC 
 
 3 
 
 »-< 
 
 c 
 
 
 © 
 
 
 
 
 oc 
 
 c\ 
 
 CM 
 
 ■^ 
 
 CO f~ 
 
 c 
 
 CN 
 
 i-< 
 
 
 
 
 
 
 
 «fl 
 
 
 © 
 
 c 
 
 o. 
 
 •-< 
 
 
 
 
 © «o 
 
 CM 
 
 CN 
 
 © 
 
 
 
 CO 
 
 o- 
 
 CO 
 
 CM 
 
 
 © 
 
 
 
 
 c 
 
 © CM 
 
 ec 
 
 
 © 
 
 
 <N 
 
 
 CM 
 
 c- 
 
 CO 
 
 ec 
 
 C 
 
 «o 
 
 
 c 
 
 
 
 
 8 
 
 
 
 
 6 
 
 1 
 
 
 
 1 
 
 c 
 
 
 e 
 
 U & 
 
 
 c 
 
 o 
 
 
 
 
 t 
 El 
 
 5. 
 
 X 
 
 J2 
 
 > 
 
 > 
 
 s-g 
 
 
 
 
 i 
 
 a- 
 
 3 
 
 a 
 
 
 
 
 
 CO 
 
 
 
 
 > 
 
 > 
 
 
 B 
 
 t 
 
 SI 
 
 hJ 
 
 l- 
 
 h3 
 
 CM 
 
 SN 
 
 CM 
 
 
 
 co 
 
 O 
 
 
 
 os 
 
 
 
 
 
 
 
 
 
 
 
 " 
 
 CM 
 
 J3 
 
 c3 o 
 
 
 
 
 
 
 
 a-. 
 
 — 
 
 © 
 
 " 
 
 " 
 
 ^ 
 
 o 
 
 
 
 o 
 
 3 
 
 3 
 
 3 
 
 
 
 
 C 
 
 C 
 
 O 
 
 X 
 
 X 
 
 X 
 
 3 
 
 c 
 
 3 
 
 C 
 
 CD 
 
 a> 
 
 s 
 
 2 
 
 s 
 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 95 
 
 CM -- ->*< CM 
 
 n ^ q oo 
 
 Tji to [<3 N 
 
 OO CO 
 
 — ' o 
 
 CI 
 
 CO 
 
 CO 
 
 co 
 
 o 
 
 >o lO 
 
 OS 
 Ol 
 
 co 
 
 CO 
 
 t— 1-1 
 
 cr. 
 
 CM 
 CC 
 
 8 
 
 OS CO 
 CM ^4 
 
 OO 
 
 oe 
 
 co 
 
 t- oo 
 
 OO lO 
 
 Ol 
 
 ~ 
 
 OS 
 CM 
 
 co -4 
 
 -r 
 
 00 
 
 CO 
 
 OS OS 
 
 co os 
 
 o 
 
 CO 
 
 — 
 
 CO 
 
 O co 
 
 CO 
 
 o 
 
 CM 
 
 CO 
 
 — < UO 
 
 ^< o 
 
 00 
 
 iO 
 
 o 
 
 CO 
 
 o -* 
 
 CM 
 
 CM 
 
 s 
 
 l~ CM 
 
 CO CO 
 OS CO 
 
 f - 
 
 i- 
 
 CO 
 
 o 
 
 
 
 o 
 
 CM CM 
 
 ©3 CO 
 
 CC 
 OS 
 
 o 
 
 co 
 
 O 
 
 
 
 o 
 
 CO CM 
 
 CO CO 
 
 CO 
 
 01 
 
 «c 
 
 00 
 
 O CM 
 
 O 
 
 CO 
 
 CO 
 
 OS OS 
 
 CO CO 
 
 o 
 ■a 
 
 CD 
 
 oe 
 
 CM 
 
 co 
 
 Oi 
 
 O CM 
 
 Q 
 
 CC 
 
 CO 
 
 oo -<f 
 
 co 
 
 S3 
 
 co 
 
 UO 
 
 CM .-H 
 
 1-1 
 
 o 
 
 o 
 
 t~- OS 
 
 O". >0 
 
 oo 
 
 CO 
 
 co. 
 oo 
 
 ■>* 
 
 CM rH 
 
 *"* 
 
 © 
 
 o 
 
 CM "<f 
 
 00 OS 
 
 I~ 
 
 OS 
 
 CO 
 
 r- oo 
 
 ■«tl 
 
 o 
 
 o 
 
 f^ CM 
 
 CO 
 
 CO 
 
 
 00 OO 
 
 •* 
 
 o 
 
 © 
 
 >J 
 
 -r 
 
 — 
 
 fl is 
 
 3 
 
 
 >, 
 
 C3 O 
 
 a) 
 
 
 ej 
 
 o -O 
 
 
 
 
 
 
 
 
 
 4) 
 
 1) 
 
 00 
 
 oj aj 
 
 
 
 
 03 
 
 - 
 
 S3 oj 
 
 3 
 
 
 
 CM 
 
 in 
 
 IN 
 
 
 CO 
 
 
 r^ 
 
 
 OS 
 
 CT-. 
 
 :-. 
 
 OS OS 
 
 ^ 
 
 ^ 
 
 CO 
 
 -fli" rH 
 
 CM 
 
 
 
 ^4 CM 
 
 ^s 
 
 
 
 
 o 
 
 c3 
 
 as 
 
 
 co O 
 
 
 
 
 
 OS 
 
 co 
 
 OS 
 
 OS OS 
 
 o 
 
 o 
 
 O 
 
 o o 
 
 d 
 
 _s 
 
 _c 
 
 •S .s 
 
 w o o o o 
 o o o o o 
 
 -a -d t3 t) 73 
 
 s s ss 
 
 CM O 
 
 o i— 
 
 <* 
 
 CO 
 
 8 
 
 ■*i CM 
 ~* CM 
 
 s 
 
 CM 
 
 o 
 
 CO o 
 CM 00 
 
 CM 
 
 CO 
 
 CO 
 
 *f CM 
 i-H CM 
 
 CO 
 
 1- 
 
 Ol 
 
 CM 
 
 CM OO 
 
 T" CO 
 
 m 
 
 Ol 
 
 CO 
 
 CO CO 
 
 CO 
 
 CO 
 
 lO 
 
 m n in 
 
 »o oo o o >o 
 
 ~ CM — 
 
 •o >o r- 
 © o o 
 
 OS CO OO .-1 O 
 N M f ■* iO 
 
 r-l -*H O O O 
 
 i^ co r>- 
 
 N O0 Ol 
 
 O O ^r 
 
 i-i o o — ' 
 
 -4 ~H o o 
 
 CO CO CO CM CO 
 
 — r-H O § fl 
 
 fl C P g >> 
 
 I* pH J CO fa 
 
 CO CO CO CO CO 
 
 co co co co co 
 oj oi oi o a 
 
 t-4 i-i i-l CO CM 
 
 lo >> >> 
 
 III 
 
 CM CM CM CM CM 
 
 5 s $ $ s 
 
 VI 'JH Tj W :jj 
 
 c3 Cj o3 c3 o3 
 
 O 00 CO 
 
 00 CO oo 
 
 ** <6 d 
 
 CM CO 
 
 CM 
 
 t^- OJ 
 ^4 CM 
 
 CO 
 CO 
 
 co co 
 
 O CM 
 
 OS 
 
 S3 
 
 «o 
 
 ■**< oo 
 
 OO CO 
 
 oo 
 
 CM 
 
 CM CO 
 CM — 
 
 CO 
 
 r- oo 
 
 t^ CO 
 
 CM 
 
 CM 00 
 
 CO 
 
 lO O 
 O -"Cf 
 
 O0 
 
 o 
 
 ^»< CO 
 
 os 
 
 Oi CO 
 
 t^ 1-1 
 
 CO CM 
 
 OS 
 
 o 
 
 o 
 
 o 
 
 CO i-t 
 CO O 
 CO CM 
 
 CO 
 
 CD 
 
 o 
 
 o 
 
 o 
 
 03 -H 
 
 ■"CP CM 
 
 oo 
 
 CO 
 
 — o 
 
 o 
 
 CO CO 
 
 co oo 
 
 CO CO 
 
 CM 
 
 " o 
 
 o 
 
 r^ O 
 
 00 
 
 CM CM 
 
 o 
 
 CM OJ 
 03 CM 
 
 OS 
 
 oo 
 
 --I CM 
 
 o 
 
 o> r^ 
 
 OS 
 
 co 
 
 co co 
 
 CO 
 
 CM -* 
 
 oo o 
 
 00 
 
 OS t- 
 
 CO 
 
 
 S S * 
 
 0) © ro 
 
 ■s ^ & 
 
 cy a> <d 
 £ fc Z 
 
 °S S o 
 
 -« ^ JaS 
 ►3 »J h4 
 
 13.36 
 14.76 
 15.62 
 
 18.61 
 12.82 
 12.83 
 
 19.21 
 16.21 
 12.69 
 
 4.92 
 10.81 
 13.15 
 
 5.69 
 10.28 
 13.49 
 
 0.264 
 
 .230 
 
 0.216 
 
 0.237 
 
 .246 
 
 0.203 
 
 1.297 
 2.486 
 2.841 
 
 1.348 
 2.529 
 2.738 
 
 1.49 
 
 1.18 
 1.11 
 
 1.61 
 1.34 
 1.22 
 
 • ^t" OS 
 
 • CO OS 
 
 • d d 
 
 6.69 
 6.98 
 7.03 
 
 S££ 
 
 3 § § 
 
 A J3 JS 
 
 o 
 
 
 
 
 
 
 
 CC 
 
 01 
 
 a 
 
 o 
 
 o ■« 
 
 OO U5 C 
 
 
 
 
 
 c<- 
 
 
 CO ^ 
 
 
 o^ 
 
 rr 
 
 
 CM CO 
 
 
 
 
 oe 
 
 
 > 
 
 o- 
 
 CM 
 
 •*f< ^H 
 
 
 
 C^ 
 
 C" 
 
 CO Tt< 
 
 CN 
 
 ■t 
 
 cr 
 
 <y 
 
 O OS 
 
 
 
 
 
 -*t< -<tl 
 
 CC 
 
 o 
 
 t- t> 
 
 CO CM 
 
 
 
 
 
 
 •** 
 
 
 o 
 
 
 
 o 
 
 CC 
 
 oe 
 
 t> 
 
 CM ■>*< 
 
 
 
 or 
 
 CC 
 
 CO CM 
 
 
 
 
 
 
 « 
 
 or 
 
 c/- 
 
 a 
 
 
 
 o- 
 
 
 K 
 
 co os 
 
 rH CO 
 
 ur 
 
 cc 
 
 Of 
 
 rr 
 
 CM i-l 
 
 
 •* 
 
 
 c- 
 
 I-H CO 
 
 
 er 
 
 o- 
 
 «fl 
 
 CO OS 
 
 
 
 
 
 
 
 
 p 
 
 CM 
 
 
 O 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 c«- 
 
 c 
 
 T 
 
 CM 
 
 ~ o 
 
 C 
 
 
 
 
 o 
 
 
 
 
 r r 
 
 CO CM 
 
 CN 
 
 i - 
 
 
 T 
 
 co r~ 
 
 ■^ 
 
 cc; 
 
 CC 
 
 
 
 c 
 
 
 
 
 o 
 
 l~ 
 
 o 
 
 
 ■** 
 
 CO CO 
 
 1 
 
 0< 
 
 CN 
 
 
 
 
 c^ 
 
 Ct 
 
 V, 
 
 
 <= 
 
 
 
 
 o 
 
 s 
 
 C 
 
 c 
 
 CC 
 
 lO CO 
 
 CM 
 
 CM 
 
 I— 
 
 C 
 
 o o 
 
 
 
 ? 
 
 CM 
 
 
 
 
 a 
 
 OS oo 
 
 CM 
 
 e> 
 
 1— 
 
 c 
 
 o o 
 
 CM 
 
 
 
 
 
 
 
 
 i/ 
 
 
 *fl 
 
 i- 
 
 — 
 
 'X 
 
 t^ oo 
 
 c 
 
 
 
 Ol 
 
 CM M 
 
 t> 
 
 c 
 
 ? 
 
 C 
 
 8 R 
 
 CC 
 
 ue 
 
 c 
 
 cr 
 
 
 C 
 
 
 CO 
 
 CM 
 
 CM ^-i 
 
 
 
 
 
 
 
 
 
 
 
 t- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 B! 
 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 q 
 
 
 
 
 
 t 
 
 i 
 
 a 
 
 OQ 
 
 
 
 = 
 
 
 > 
 
 
 a 
 
 
 
 t 
 
 
 
 
 ; r* 
 
 
 T) +^ 
 
 
 - 
 
 
 > 
 
 
 er 
 
 v 
 
 b 
 
 
 oc 
 
 1 
 
 c 
 
 ■i 
 1 
 
 c 
 c 
 
 I 
 
 DC 
 
 II 
 o <u 
 
 C 
 
 ^c 
 
 ^ 
 
 QQ 
 
 i« 
 
 >c 
 
 IC 
 
 iej 
 
 UD «3 
 
 
 
 
 
 
 o 
 
 cr 
 
 a 
 
 
 
 
 
 
 
 3 8 
 
 
 
 c 
 
 ec 
 
 
 
 ^ 
 
 
 
 c 
 a 
 i-s 
 
 
 1 
 
 1 
 
 ■ll 
 
 
 
 
 •»f 
 
 •^fc -^ 
 
 
 
 
 
 co co 
 
 aa 
 
 cr 
 
 a- 
 
 cr 
 
 
 0? 
 
 o- 
 
 o- 
 
 & 
 
 
 
 
 
 
 b 
 
 s 
 
 a 
 
 J 
 
 03 § 
 
 J3 
 
 ,X 
 
 ^ 
 
 SX -d 
 
 CC 
 
 CC 
 
 DC 
 
 CC 
 
 CO 
 
 aj 
 
 
96 
 
 University of California — Experiment Station 
 
 
 
 
 
 <f M ^ 
 
 
 
 ^ t> 
 
 
 co •«« — 
 
 
 
 
 
 
 CO OO t> 
 
 
 © CM O 
 
 
 OO rt T(| 
 
 
 -ofe 
 2-° 
 
 paujnqufj 
 
 
 *4 «*" oo 
 
 CM CM CT 
 
 
 ffl N C 
 
 > 
 
 1" to « 
 
 r-( ^H CM 
 
 
 OS — 1 oc 
 
 
 ^H l^ « 
 
 
 ^l N N 
 
 
 
 
 o^ 
 
 
 
 in « is 
 
 
 00 CO c 
 
 
 t^. it< CO 
 
 
 paujng 
 
 
 CO d t- 
 CM CM •<* 
 
 
 t>" en *■ 
 
 .-I rt c 
 
 
 CO OO CM* 
 
 ^H ~H CO 
 
 
 t^ CO — 
 
 
 r>. oj ce 
 
 
 r-H CM CO 
 
 
 
 
 
 
 
 
 o> cm r- 
 
 
 OJ © O 
 
 
 00 o »-< 
 
 
 ®.S 
 
 paujnqufl 
 
 
 -o" ■*£ cn 
 
 CM !-H 
 
 
 •"l" r>" •*» 
 
 CM »-c 
 
 
 a s * 
 
 
 "0 a> 
 3 •o* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 E. o 
 
 
 
 ffl »h 
 
 
 ■* N ^> 
 
 
 
 
 OS. 
 
 
 
 CO OO CC 
 
 
 OO CO C\ 
 
 
 CO CO CM 
 
 
 paujng 
 
 
 W N o 
 CM n 
 
 
 "O CJ> If 
 
 
 Ifl CM* Tf 
 CM CM 
 
 
 
 
 H M CT 
 
 
 if; t^ oc 
 
 
 lO to * 
 in c» co 
 
 
 
 
 
 CO CO if 
 
 
 CNJ oo «■ 
 
 
 
 c3 o3 
 
 paujnqufj 
 
 
 .-" I-H C" 
 
 
 ex 
 
 
 oc" t-* o" 
 
 
 OJ -tf< OC 
 CO «* w 
 
 
 CO CO l> 
 co •>»< o 
 
 
 CO o o 
 
 
 
 
 paujng 
 
 
 ,-" ,-" ** 
 
 
 cd od t^ 
 
 
 00 00 o 
 
 
 CD •«»< «- 
 
 
 U0 OJ »- 
 
 
 CO ^t CO 
 
 
 OR 
 
 3 
 o 
 
 
 
 
 
 CNN if 
 
 
 CO O l> 
 
 
 00 1-H ifj 
 
 
 paujnquQ 
 
 
 T* CM C= 
 
 © " c 
 
 
 -c* co c 
 © " c 
 
 
 CO CO r-l 
 
 o * d 
 
 
 .3 
 
 a 
 
 
 
 
 
 
 
 
 
 ^h — i e<- 
 
 
 » N C 
 
 
 S H to 
 
 
 
 
 
 
 
 3SS 
 
 d c 
 
 
 o c 
 
 
 CO CM CO 
 
 
 
 paujng 
 
 
 
 
 CO CO -^ 
 
 d ' d 
 
 
 H (O « 
 
 
 o> t^. ■*» 
 
 
 o> O CO 
 
 
 
 
 
 
 
 
 <D O! OC 
 
 
 CO CM 1> 
 
 
 CO lO t 
 
 
 _3 
 
 "3 
 
 *c3 
 
 paujnqufl 
 
 
 lO (N r- 
 
 © ' c 
 
 
 I->. xj< CC 
 0» <M* t- 
 
 
 CM CM* i- 
 
 
 
 <M ■*»< t- 
 
 t^ o o- 
 
 
 CO -H CN 
 
 •^ CO CC 
 
 
 
 
 
 
 N CO tD 
 O lO CO 
 
 
 O 
 
 paujng 
 
 
 If) M - 
 
 d ' c 
 
 
 OO to h 
 <m' cm' — 
 
 
 CO cm" ^ 
 
 
 
 CO CO t- 
 
 
 M CO U 
 
 
 
 
 
 
 
 t^ CC •»»< 
 
 
 a 
 
 .2 
 
 
 
 » «5 OC 
 
 
 
 Tf M N 
 
 
 paujnqufl 
 
 1 
 
 a 
 
 N IN C 
 
 
 CO CM <- 
 
 
 CO CO H 
 
 
 OO OO OC 
 
 
 cm -« a 
 
 1 
 
 r-i i-H OO 
 
 
 
 
 
 
 3 
 
 t^- »fj OC 
 
 
 «) O) c 
 
 OO •>»< CO 
 
 "TS 
 
 o 
 
 paujng 
 
 o 
 
 N N C 
 
 S 
 
 00 
 
 CO N - 
 
 a 
 
 CO CO ^h 
 
 s 
 
 
 
 i 
 
 
 
 
 
 
 co OO C 
 
 OJ I>- «C 
 
 s 
 
 CO ■>»< — 
 
 J 
 
 CM ■«*< oi 
 
 
 
 
 
 CD 
 
 
 V 
 
 00 O C 
 
 
 o o io 
 
 o 
 
 e3 co 
 
 paujnqufj 
 
 >> 
 a 
 3 
 
 (D «5 K 
 
 >> 
 
 CN) CM OC 
 
 
 n h to 
 
 o 
 
 ■*r r~ — 
 
 © 
 
 -H CO — 
 
 o 
 
 CO CM -h 
 
 
 
 
 
 OO O f 
 
 
 CM CM If 
 
 60 
 
 
 1 
 
 fc 
 
 paujng 
 
 1 
 
 02 
 
 CO Ifj CO 
 
 
 CO N N 
 
 3 
 
 w 
 
 CO -h to 
 
 b- 
 
 
 <i 
 
 
 co 
 
 
 I— 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 go 
 
 
 
 
 ._ 
 
 
 
 
 3 
 
 
 
 
 
 H 
 
 
 
 00 
 
 01 
 
 
 
 
 3 
 
 
 
 
 O 
 
 
 
 
 
 J 
 
 
 
 A 
 
 
 
 
 o 
 
 
 
 
 •M 
 
 
 
 
 
 tt 
 
 
 o 
 
 e 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 < 
 
 
 a 
 
 
 
 
 T2 
 
 
 
 
 13 
 
 
 
 
 
 
 I 
 
 2 
 
 .a 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 .rj 
 
 73 
 09 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 | 
 
 ti 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 2 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 a 
 
 03 
 
 
 
 bO 
 
 
 
 
 a, 
 f. 
 c 
 
 
 
 
 
 
 
 
 o 
 
 
 
 c 
 
 03 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 T3 
 
 
 
 
 
 
 DO 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 Eh 
 
 
 
 
 
 
 
 
 
 
 
 c« 
 
 
 CD 
 
 be 
 
 3 
 
 
 6 
 
 
 
 9> 
 
 a 
 
 
 
 1 
 
 
 5 
 
 GO 
 
 
 
 bfi ■ oc 
 
 sj a eS 
 
 
 be o 
 
 
 
 
 Young leaf 
 Full bloom 
 Drv. seeds 
 
 
 Early leaf s 
 In full bloo 
 
 
 Early leaf s 
 In full bloo 
 Nearly dry, 
 
 
 
 
 «3 W lO 
 
 
 CO co CO 
 
 
 co co co 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i^ M 
 
 
 
 
 
 
 
 
 
 o.g 
 
 
 
 
 
 
 
 
 
 
 
 CO CM* <M* 
 
 
 * tDN 
 
 
 * tO N 
 
 
 
 -2 'a 
 
 
 <M ^H 
 
 
 CN 
 
 
 CM 
 
 
 
 02 fl 
 
 10 
 
 
 ,3 
 
 
 A 
 
 
 ^ _ 
 
 
 
 
 d 2 > 
 
 g S Cj 
 
 
 Marc 
 April 
 June 
 
 
 S a § 
 
 
 
 
 
 £s s 
 
 
 
 S«i ^ 
 
 
 
 
 , T3 
 
 
 
 
 
 
 
 
 
 u. CD 
 
 
 
 
 iO »fj lO 
 
 
 "5 "5 if) 
 
 
 
 S ? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 * 3 
 
 
 
 
 
 
 
 
 
 .£) 
 
 
 
 
 
 
 
 
 
 vS'S 
 
 
 
 o o o 
 
 
 COO 
 
 
 
 a' O 0) 
 
 
 
 
 .S .S .£ 
 
 
 .s .s a 
 
 
 
 ^1 
 
 
 Cj Cj 6 
 
 
 'S '3 "S 
 o o o 
 
 
 "3 *S 'S 
 
 o o o 
 
 
 c 
 
 5 "^ S 
 
 
 
 
 -O -0 -D 
 
 
 T3 T3 T3 
 
 
 C 
 
 >.S8 
 
 
 3 .3 -C 
 
 
 3 3 fl 
 co o a 
 
 
 3 3 3 
 S 4) 1) 
 
 
 
 
 
 CO 
 
 CO 
 
 /. 
 
 
 S 
 
 <*, 
 
 <- 
 
 
 § 
 
 («! 
 
 <^ 
 
 
J Jul. 085] Plant Succession on Burned Chaparral Lands 97 
 
 by hand, whereas those of the herbs included the entire top growth. All sam- 
 ples were treated in the same manner in their preparation for analysis. 25 
 
 Table 17 gives information on sampling, and the results of the analysis. 
 
 The crude protein content is seen to be somewhat higher in the plant samples 
 collected on the burned areas, from the early growth stage until after flower- 
 ing. This observation agrees with the results of other workers. Hart, Guilbert, 
 and Goss (32) found certain species to contain a higher crude protein content 
 on burned California chaparral areas. Aldous (1) reported the percentage of 
 crude protein of bluestem grasses to be slightly higher in early June in the 
 regenerated growth on areas burned late in the spring. Wahlenberg, Greene, 
 and Reed (113) partly support this point for burned pine lands of the South. 
 Neal and Becker (71) found the samples of wire grass taken from burned 
 areas to have a higher crude protein content than those collected from adja- 
 cent unburned areas. 
 
 In the late growth stages, the data in table 17 show no consistent difference 
 in the percentage of crude protein in the two groups of samples. Both the shal- 
 low-rooted grasses and the broad-leaved herbs show greater increases of crude 
 protein when growing on burned areas than do the deeper-rooted shrubs. This 
 may partly be accounted for by the addition of ash and the increased nitrifica- 
 tion in the surface soil layer of the newly burned areas (see page 83) . 
 
 The crude fiber content shows higher values in the plants collected on the 
 newly burned plots, notably in the intermediate growth stage. Burning has the 
 least effect on the crude fiber content in two nondeciduous shrubs, chamise and 
 whiteleaf manzanita, but even in these species the content is slightly higher in 
 the samples collected on the burned areas. The relatively larger, heavier leaves 
 and the longer, coarser, more woody stems of the samples collected on the 
 recently burned lands would appear to account for the percentage differences 
 in the crude fiber content of these species. 
 
 The mineral constituents (table 17) showed no consistent differences within 
 specific growth stages, in either the shrubs or herbs grown on burned or 
 unburned areas. In all species the levels of phosphorus and potassium are 
 highest in the earlier growth stages. 26 The percentage of calcium, on the other 
 hand, increases in the shrubs and decreases in the herbs with the march of the 
 season. Thus in the fall, in most of the chaparral and other woody species 
 studied, the ratio of calcium to phosphorus is conspicuously high, an indica- 
 tion of an undesirable balance from a forage viewpoint (68) . 
 
 25 The fresh weights of the samples were recorded in the field immediately following the 
 collection, after which the material was dried in an electric oven, with circulating air, at a 
 temperature of 70° C. When dry, the samples were finely ground, bottled, and tightly stop- 
 pered. After the analyses of the samples were completed, an aliquot part of each sample 
 was dried at 100° C to constant weight; the analytical data were then expressed on this 
 moisture-free basis. 
 
 In the determination for nitrogen, calcium, ash, and moisture, Official and Tentative 
 Methods of Analysis of the Association of Agricultural Chemists (4) was followed. Phos- 
 phorus was determined colorimetrically, as described by Kuttner and Cohen (57). Crude 
 fiber was determined by the method described by Sharrer and Kurschner (94). Silica was 
 determined by treating the ignited residue of the sample with concentrated HC1 and 60 
 per cent HC10 4 , igniting the residue after digesting on the steam bath, and then filtering. 
 
 26 It is not surprising, perhaps, that this constituent does not show higher values even in 
 the shallow-rooted annual species, since the phosphorus liberated by burning is readily 
 fixed by the soil complex, or is precipitated as insoluble ferric phosphate, or, in the pres- 
 ence of excessive calcium and a high pH, forms calcium phosphate. 
 
98 University of California — Experiment Station 
 
 Although burning apparently affects but little the chemical composition of 
 individual species at a specific stage of growth, the change in cover and the 
 frequently greater vegetative growth resulting from burning do increase 
 the total nutrients on the area, and therefore favor greater nutritional in- 
 take of the grazing animals. 27 The abundant, rapidly growing, succulent shoots 
 of the browse plants, with their relatively large leaves, are within easy reach 
 of the animals ; and, as pointed out, there is a slightly prolonged period of suc- 
 culence of the vegetation as a whole. 
 
 FIELD OBSERVATIONS OP AREAS BURNED UNDER 
 STATE PERMIT 
 
 In the preceding pages detailed measurements and study of plant succes- 
 sion, forage yield, soil reactions resulting from burning, and chemical values 
 of the forage have been reported. In the discussion which follows, based on 
 field observations, the broader trends resulting from burning of a large num- 
 ber of diversified areas are presented. The points recorded are essentially those 
 which determine the value of the burning venture. 
 
 In 1938 several privately owned chaparral and chamise areas in the north- 
 ern counties of the state were burned under the sanction and supervision of 
 the State Division of Forestry. Seven weeks during the summer and autumn 
 of 1939 were devoted by the writer and a field assistant to a critical review 
 of these supervised burns and several others, 34 in all, in an effort to evalu- 
 ate the degree of success attained by burning. Although close comparison with 
 the carrying capacity before burning of each specific area was not possible, the 
 vegetation on similar adjoining unburned areas was noted and compared. 
 Most of the areas had been burned one to two years prior to the examination. 
 A field observational outline guide was used in recording the conditions. The 
 points noted included such facts as general history of the area, topography, 
 soil type, character and density of the vegetation before and after burning, 
 extent of utilization, degree of erosion, and estimated carrying capacity. 
 
 The degree of soil erosion was classified as "normal to light," "moderate," or 
 "severe." Normal to light soil erosion implied no appreciable departure from 
 the geologic rate, and was characterized by the presence of fragments of 
 charred wood scattered over recently burned areas, the occurrence of rounded 
 out and revegetating older gullies, and in some instances by a well-developed 
 top soil which contained abundant organic matter. Severe soil erosion was 
 recognized by the presence of numerous new or well-defined gullies, most of 
 which were V-shaped and which were revegetating little if at all, by exposure 
 of dark areas on the crowns of chaparral denoting current removal of soil by 
 general sheet erosion, and sometimes by the presence of soil deposition on 
 drainage areas below. Moderate soil erosion was recognized as being in the 
 early stages of accelerated soil movement, and was indicated by the presence 
 of a few incipient gullies, and of considerable sheet erosion. 
 
 The grazing-capacity estimates, always difficult to derive, were obtained in 
 part by taking into account the period of time that a known number of animals 
 
 27 Owe, John L. The effect of nitrification upon vegetation and forest reproduction in some 
 California pine forests. Thesis for the degree of Master of Science, University of California, 
 1930. Copy on file in the Library of the University of California, Berkeley. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 99 
 
 actually grazed upon the area, the reaction of the range to this stocking, and 
 partly by the employment of established grazing reconnaissance computa- 
 tions. Forage requirements were computed on the basis of 0.6 of a forage acre 
 per cow month, or per animal unit per month. As here used, animal unit is the 
 acreage required to provide one cow, or its equivalent of 5 sheep, with this 
 amount of forage. The term "stand" is used to indicate the amount of ground 
 covered by vegetation, and is expressed in per cent. Allowance was made for 
 the forage values of plants which are in evidence only during the spring pe- 
 riod. The following discussion summarizes, by counties, the successional and 
 other important information recorded during the inspection of these areas. 
 
 Lake County. — Area 1 : The original vegetation on this 150-acre area, 
 burned in 1936, consisted of a dense stand of pin oak 2 to 12 feet in height. At 
 the time of inspection, two years after burning, pin oak sprouts were rapidly 
 replacing the annual grasses and weeds, which then had a density of 20 per 
 cent. The area had been grazed by sheep and cattle; about 6 acres was 
 required per animal unit per month. No artificial reseeding had been 
 attempted. The stony loam soil of the Los Osos series of this steep area, averag- 
 ing 50 per cent slope, varied from bare rock to 8 inches in depth, and had dried 
 hard on exposed areas. Soil erosion was light, and was chiefly confined to 
 spotty sheet erosion, and to the formation of a few incipient gullies. 
 
 On the portion of this ranch where the cover consisted primarily of chamise, 
 greenleaf manzanita, canyon live oak, and annual grasses, the vegetation had 
 been so closely browsed that much of the brush was killed before the area was 
 burned. The vegetation at the time of inspection consisted of yerba santa, 
 Junegrass, foxtail fescue, wild oat, Napa star thistle, and chamise. 
 
 Area 2 : Part of this area, consisting of 500 acres, was burned in 1938, and 
 was unusual in that in some units the chamise sprouts and seedlings, and some 
 other brush species, were so closely browsed by deer since burning as to open 
 up the cover conspicuously. Little herbaceous vegetation had appeared when 
 the area was inspected, but the chamise and deerbrush comprised a density of 
 15 per cent. No stock was grazed on the newly burned unit owing to the poor 
 forage conditions. The soil, a stony clay loam of the Hugo series, varied from 
 surface rock to 3 feet in depth. The average slope was 10 per cent. Soil erosion 
 was light, and was essentially limited to sheet erosion on the more exposed 
 slopes. Other areas on this ranch had been previously cleared of brush by the 
 use of fire through a planned and continuous long-time burning program. 
 Only the more gentle slopes which embrace the deeper, more fertile soil had 
 been burned. These clearings supported the following species : foxtail fescue, 
 wild oat, red brome, silver hairgrass, nitgrass, needlegrass, Napa star thistle, 
 tarweed, and St. Johnswort. Wild oat had been seeded artificially with some 
 apparent success. Success in clearing of the brush, and the invasion of an 
 annual herbaceous cover, were particularly impressive. The grazing capacity 
 was fairly high, and the burning had been economically successful. 
 
 Area 3 : In 1936 the heavy stand of chamise was completely burned on this 
 150-acre unit. At the time of inspection, in 1939, a cover of chamise sprouts 
 and seedlings of about 40 per cent density had recaptured the area, with 
 almost no other vegetation in evidence. There was so little forage that 9 acres 
 was estimated as required per animal unit per month ; hence the area had 
 
100 University of California — Experiment Station 
 
 practically no pasturage value. The soil, a gravelly, sandy clay loam of the 
 Konokti series, varied in depth from surface rock to IV2 feet. The average 
 slope was 5 per cent. No abnormal soil erosion was indicated. The owner stated 
 that he had burned this area not with the hope of procuring a grass cover, but 
 merely to have the use of the browse for a short time in the spring. 
 
 Area 4 : In 1938 a chamise area of 150 acres which adjoined area 3 was 
 heavily burned. When inspected, it was virtually devoid of vegetation other 
 than chamise seedlings and sprouts of about 25 per cent density, the latter 
 being short and bushy as a result of their having been closely browsed by deer. 
 This area was pastured neither before nor after burning. In its state of low 
 forage growth when inspected, this burn could carry no stock economically, 
 unless it be merely to provide an occasional menu of chamise to serve as variety 
 with the grass diet from nearby lands. The gravelly, sandy loam of the Ko- 
 nokti series varied from exposed rock to 1% feet in depth, and was unusually 
 stony. The average slope was 3 per cent. There was little evidence of abnormal 
 erosion. 
 
 Area 5 : Originally supporting a dense stand of California scrub oak and 
 Stanford manzanita, this 200-acre tract was burned in 1938. The area had not 
 been grazed in 1939 when inspected ; in fact, there was almost no vegetation 
 except scrub oak sprouts and scattered plants of St. Johnswort (89). With a 
 plant cover of only 15 per cent density, the carrying capacity was obviously 
 too low to utilize for domestic grazing animals. Nevertheless the soil, a sandy 
 loam of the Konokti series, showed only normal erosion, with some sheet ero- 
 sion on some of the steeper facings. The average slope was 10 per cent. 
 
 Mendocino County. — Area 6 : The rapid return of the mixed chaparral was 
 distinctly conspicuous on this 15-acre burn, despite the fact that the fire of 
 1938 resulted in removing the top growth of nearly all the old stand. Over- 
 utilization of the area in 1939 had resulted in leaving little of the herbaceous 
 vegetation, even of species of secondary palatability. Although the area 
 was pastured only in the spring, when the forage was at its best, estimates 
 indicated that 7 acres was required per animal unit per month. The average 
 slope for the area was 15 per cent, whereas the steeper slopes averaged 42 per 
 cent. The gravelly, sandy clay soil, of the Mariposa series, showed severe ero- 
 sion on the steeper slopes. The erosion ranged from small gullies to slips of 
 large blocks of soil. Before burning, the area was occupied by oak woodland, 
 with some redwood present. Following burning, browse furnished limited feed 
 the first year, whereas grass composed the bulk of the forage during the second 
 and third years. The owner stated that reseeding or cultivation of any sort was 
 impossible because of the presence of the numerous, extensive gullies. 
 
 Area 7 : Burning some five times during the last twenty years was the his- 
 tory of this 140-acre area, with the last fire occurring in 1935. Wavyleaf cea- 
 nothus and greenleaf manzanita, which formerly predominated, formed a den- 
 sity of about 35 per cent, and were rapidly reoccupying the site. Herbaceous 
 vegetation, composed primarily of foxtail fescue and Scouler St. Johnswort, 
 was so sparse that the soil was nearly devoid of an understory cover. Grazing 
 values were so low as to be little or nothing, about 8V2 acres being required per 
 animal unit per month. The average slope was 50 per cent. The unprotected 
 so r face soil showed the pebbly pavement common to the action of sheet ero- 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 101 
 
 sion, whereas a few soil slips, associated with prominent fresh gullies, indicated 
 active but moderate soil erosion. 
 
 Area 8 : After a heavy stand of redwood timber was logged, a particularly 
 dense, tall growth of blue blossom took over this entire tract. The woody 
 growth on 300 acres was lopped in 1936, and burned late in the autumn of 
 1937, after the vegetation had dried. For the past several years 50 acres or so 
 had been cut and burned each year, until some 400 acres had been more or less 
 temporarily cleared, except for the extensive sprouting of the numerous red- 
 wood stumps and scattered stands of various brush plants. The lopping and 
 burning cost from $7 to $10 an acre. After one or two heavy rains, the recently 
 burned areas were seeded by the owner to a mixture of redtop, timothy, peren- 
 nial ryegrass, velvet grass, Kentucky bluegrass, and orchard grass at the rate 
 of 20 pounds to the acre, and at a cost for the seed of about $2.50 per acre. 
 Also, in the earlier operations, mustard seed was included with the mixture at 
 the rate of 3 pounds to the acre. This seeding resulted in the establishment of a 
 cover of from 20 to 50 per cent for the first and second years after seeding. In 
 the third year, however, annual grasses, weeds, California huckleberry, Pacific 
 madrone, and blue blossom had all but crowded out the cultivated plants. 
 Because of this failure, Harding grass was later substituted in the reseeding 
 endeavor on the chopped and subsequently burned areas ; it made a much 
 better showing than did the other grasses. A good stand of Harding grass was 
 obtained for the first three years or so after seeding, but later on this grass 
 tended to wane ; yet it was clearly the most stable species introduced. Although 
 artificial reseeding had been partially successful, approximately 6 acres was 
 nevertheless required per animal unit per month. The area has been grazed by 
 sheep. Aside from the encroachment of the brush and redwood sprouts, exces- 
 sive yearlong grazing mostly accounted for the rather limited forage supply. 
 The average slope was 25 per cent. Recent erosion of the light Melbourne sandy 
 clay loam soil had been severe on many of the steeper slopes, and was reflected 
 in the low productivity of the site. Soil slip, with prominent fresh gully forma- 
 tion below, was particularly impressive on several of the steeper, earlier- 
 cleared slopes. It is significant, perhaps, that this ranch changed hands three 
 or more times in the past ten years, but following each of these transactions the 
 property reverted to the original owner. 
 
 Area 9 : This 10-acre area, located in a nearby level swale, was of superior 
 site quality and until recently, when the area was logged and burned, was 
 occupied by redwood trees 10 to 60 inches in diameter. The average slope was 
 20 per cent. The soil was deep, and varied from friable loam to rocky clay. 
 There was no evidence of abnormal soil movement or loss. Many redwood 
 sprouts 4 to 34 inches in height, and many seedlings of blue blossom, were 
 reinvading this burned area. Despite reseeding in 1938 to various cultivated 
 grasses, annual volunteer species of grasses, thistles, and many weeds clearly 
 predominated, and formed a density of 45 per cent. The area was lightly 
 grazed by sheep in the spring of 1939, following which herbaceous growth of 
 fair to good quality was produced. 
 
 Area 10 : The general soil conditions of this 4,000-acre burn appeared seri- 
 ous. Erosion of one form or another was severe and was taking place at a rapid 
 rate. The average slope was 63 per cent. Soil slip was particularly impressive 
 
102 University of California — Experiment Station 
 
 in that area ; on many slopes, areas of 2 to 5 acres had moved down hill ; most 
 of these slides occurred slightly below the summit on the steeper slopes. Al- 
 though there were a few small grassy flats, most of the topography was excep- 
 tionally rough for grazing ; many slopes had a 65 per cent gradient, and a few 
 more than 90 per cent. The clay loam soil of the Hugo series varied in depth 
 from a few inches to 3 feet, with considerable rock outcrop on the more prom- 
 inent ridges and slopes. A large acreage had been cleared of brush by repeated 
 burning, but at a great sacrifice of the top soil. The exposed subsoil was heavily 
 baked when examined. No charred wood remained on the slopes of recent 
 burns. The density of the vegetation over the steeper country did not exceed 
 an average of about 10 per cent, whereas it attained a density of 20 to 35 per 
 cent on the gentler slopes and flats. Because of the inferior grazing value of 
 the dominant herbaceous species, composed chiefly of foxtail fescue and nit- 
 grass, 6 acres per animal unit per month was estimated to be the range require- 
 ment. Range use at the time of inspection was clearly excessive. Numerous 
 reseeding trials, made by the owner, had failed. 
 
 Area 11 : This 60-acre chamise area was burned three times within eight 
 years, the last burn being in 1936. The area had been largely cleared of brush, 
 but the present herbaceous vegetation, with a density of approximately 20 per 
 cent, had been severely overgrazed. There was little palatable browse. Some 7 
 acres was required to maintain an animal unit for 1 month. The average slope 
 was 47 per cent. The soil, a Pinole gravelly alluvium, without clay subsoil, had 
 eroded severely, and the clearing of the brush had been accompanied by heavy 
 soil loss. There had been much soil slippage, subsequently accompanied by the 
 formation of many prominent gullies. Sand, gravel, and debris had washed to 
 the lower levels and had been deposited along the main drainage channel. 
 
 Area 12 : The original dense stand of this 60-acre chamise area was burned 
 in 1938. When examined, there was a scattering of foxtail fescue, and a rather 
 heavy stand of deerweed, together with an abundance of chamise seedlings 
 and sprouts. For the entire area burned, the density of vegetation averaged 
 20 per cent. The gradient averaged 50 per cent, and sheet and gully erosion 
 were pronounced on the upper slopes. It was estimated, exclusive of the lim- 
 ited browse value of the chamise, that 10 acres would be required per animal 
 unit per month. In the season after the fire, sheep grazed the area for a short 
 time in the spring. Soil erosion was moderate. Charred wood, ashes, and some 
 soil had been deposited on lower levels by surface runoff. 
 
 Area 13 : This 130-acre chamise area, with an average of 25 per cent slope, 
 supported vegetation similar to that of area 12. The chamise was approxi- 
 mately 3 feet in height and occupied about one fourth of the ground cover. 
 The herbaceous vegetation, with a density of 60 per cent, included several 
 superior native forage grasses. The forage was far superior to that on area 12, 
 the grazing requirement being about 4 acres per animal unit per month. The 
 average slope was 35 per cent. Soil erosion was classed as light. Although this 
 burn contained a large soil slip of recent origin, and several large gullies, no 
 portion of the area appeared recently to have eroded abnormally. 
 
 Area 14 : The vegetation of this 1939 early summer burn of 35 acres con- 
 sisted essentially of young" chamise sprouts a few inches in height, and of 
 numerous chamise seedlings. Although no reseeding was attempted, the owner 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 103 
 
 expected, in the spring of 1940, to get some feed for a few animals from the 
 sprouts. The average slope was 25 per cent. Most of the charcoal had been 
 removed by surface runoff, and moderate current sheet or gully erosion had 
 taken place. On a small portion of the area, which was dominated by interior 
 live oak, were found purple needlegrass, red brome, and blue wild-rye grass, 
 these forming a density of 10 per cent. Because of the limited forage, no 
 attempt was made to graze the area in 1939. 
 
 Area 15 : The madrone sprouts on this four-year-old burn of 100 acres were 
 of such density that it was difficult to walk through them. The few areas which 
 were sparsely occupied by brush supported a scattered stand of herbaceous 
 vegetation. Wisely, no livestock were placed on the area in 1939 because of the 
 inferior forage. The average slope of this unit was 8 per cent. No abnormal soil 
 erosion was noticed, but the soil was heavily baked. An adjoining unburned 
 area of Douglas-fir, madrone, and interior live oak was practically devoid of 
 an understory of brush, but supported considerable grass, which gave it a 
 parklike appearance. This area afforded much more grazing than did the 
 burned area, and was being utilized effectively by sheep. 
 
 Area 16 : On this 80-acre two-year-old burn chamise and manzanita were 
 coming in strongly, but heavy grazing appeared to have reduced the her- 
 baceous vegetation so that it formed only 10 per cent of the ground cover. The 
 average slope was 20 per cent. Despite the fact that the light-textured soil had 
 been severely trampled, there had been only moderate sheet and gully erosion. 
 Apparently the whole ranch had been subjected to heavy overutilization. On 
 the burned area the grazing requirement was estimated as some 8 acres per 
 animal unit per month. 
 
 Humboldt County. — Area 17 : After a young stand of Douglas-fir had been 
 slashed on this 100-acre unit, and subsequently burned, the area had been fired 
 on an average of every four years for twenty years in an unsuccessful effort 
 entirely to kill the manzanita, blue blossom, and poison oak. Each burning was 
 followed by seeding with perennial ryegrass, but this effort had proved un- 
 economical. Generally, a good stand of this grass was obtained the first year 
 after seeding ; a scattered cover remained the second season, whereas almost 
 none was to be found thereafter. The pasture was grazed lightly by sheep 
 throughout the year. About 3 acres was required per animal unit per month. 
 The soil, a stony loam of the Melbourne series, varied from 1 to 3 feet in depth. 
 The average slope was 18 per cent. Soil erosion was light ; there was but little 
 soil slip, and there were only a few small gullies. The erosion appeared not to be 
 of recent origin, since the scars and rounded-out gullies were reasonably well 
 healed over with vegetation. The clearing of this area was highly successful. 
 
 Area 18 : This 300-acre area, last burned in 1935, was predominantly 
 Douglas-fir, Pacific madrone, and tanoak, much of which had been replaced by 
 grass. After cutting and burning, the area was reseeded to perennial ryegrass 
 and orchard grass. Although both grasses became established, the latter gave 
 the better results, notably on the north slopes, where, under deferred grazing 
 and moderate stocking, it had produced a fair to good cover for four years and 
 appeared to be strong. The practice had been to graze the area lightly in the 
 spring and summer by sheep and cattle, with the result that only iy 2 acres 
 was required per animal unit per month. The average slope was 12 per cent. 
 
104 University of California — Experiment Station 
 
 The stony loam soil of the Melbourne series varied from surface rock to 3 feet 
 in depth and showed no abnormal erosion. The soil was mellow, and there was 
 little evidence of baking of the slightly exposed soil surface. 
 
 Area 19 : In 1938 the Douglas-fir, Pacific mad rone, and tanoak on this 100- 
 acre area were cut and burned ; then, in late autumn, came reseeding to peren- 
 nial ryegrass, little of which was in evidence at the time of inspection. Small 
 areas had been alternately burned each year in an effort to destroy remnants 
 of the original woody cover, and this operation had been followed by artificial 
 reseeding with cultivated grasses. The introduced grasses had failed to become 
 established, according to the owner, because of heavy grazing. The herbaceous 
 cover had a density of 20 per cent, but it was rapidly giving way to bracken 
 fern and to the original woody vegetation. Only 2% acres was estimated as 
 being required per animal unit per month. The average slope was 25 per cent. 
 The stony loam soil had eroded but lightly, and there was only limited baking 
 of the exposed soil. 
 
 Area 20 : The California hazelnut brush, Douglas-fir, and tanoak of this 300- 
 acre coastal ranch, located about a half mile from the ocean, was burned in 
 1938 and was reseeded in the autumn to perennial ryegrass and ribgrass. The 
 rather heavy forage at the time of inspection consisted of perennial ryegrass, 
 bluegrasses, Junegrass, English plantain, and thimbleberry. The long period 
 of succulence and the relatively heavy yield of this herbage appeared to be 
 accounted for by the fact that fog from the ocean provided moisture at critical 
 periods. The 60 per cent density of vegetation, and the light spring and sum- 
 mer grazing, partly accounted for the low requirement of 1 acre per animal 
 unit per month. The slope averaged 10 per cent. The rich clay loam soil, of the 
 Melbourne series, varied from 2 to 6 feet in depth, and showed little abnormal 
 erosion of any form. 
 
 Area 21 : This 100-acre area was similar in native cover, soil, and topog- 
 raphy to the one just discussed. It likewise was burned in 1938 and was 
 seeded to perennial ryegrass in the autumn of that year. In common with other 
 stockmen in the area, the owner slashed and permitted the brush to dry before 
 burning, but left the Douglas-fir stand to be killed by the fire. There was a 
 herbaceous cover of about 40 per cent density when the area was inspected, 
 and only 1 acre per animal unit per month was required for moderate spring 
 and summer grazing. Unless the area was reburned every three or four years, 
 tanoak sprouts, thimbleberry, and bracken fern tended to recapture the soil 
 to the virtual exclusion of other vegetation. The average slope was 10 per cent. 
 The loam soil, of the Melbourne series, varied in depth from 2 to 6 feet. Neither 
 erosion nor baking of the soil was abnormal. 
 
 Shasta County. — Area 22 : This 100 acres of relatively level brush and 
 woodland, burned in 1936, for the most part occupied a ridge top. Parry man- 
 zanita and canyon live oak produced a dense stand before burning, and the 
 understory vegetation then consisted chiefly of foxtail fescue and red brome. 
 When inspected, the invading vegetation formed a density of only about 15 
 per cent, but even so the fire had resulted in effectively opening up the brush. 
 Canyon live oak sprouts and manzanita seedlings, however, were clearly re- 
 gaining possession of the ground. The area had been grazed by cattle for a few 
 weeks each spring. Five acres was required per animal unit per month. The 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 105 
 
 average slope was about 20 per cent. The grayish-red stony loam soil showed 
 only light sheet erosion, and there was no serious baking of the soil. 
 
 Area 23 : The extensive chaparral flats east of Redding are fairly well char- 
 acterized by this 70-acre burn, as well as by that designated as area 24. Dwarf 
 interior live oak predominated throughout. The area was heavily burned in 
 1937. The soil was thin, with much outcrop in evidence. The average slope was 
 15 per cent, and current erosion was light. Prior to burning, only the openings 
 supported any appreciable amount of herbaceous vegetation. The second year 
 after burning, a mass of woody sprouts and seedlings had occupied the area, 
 and primarily consisted of dwarf interior live oak, blue oak, and poison oak ; 
 this invasion of chaparral had also extended into many former openings. The 
 various annual grasses, with a fair sprinkling of Spanish-clover, were so 
 sparse as to afford little grazing, and were of value only early in the spring, 
 when a few cattle utilized the area. Five acres was required per animal unit 
 per month. Considerable sheet erosion took place the first year after burning, 
 as indicated by measurement with erosion stake levels, and as evidenced by the 
 pebbly soil surface. Infiltration studies showed that water percolated into the 
 soil much more slowly than on a similar adjacent, long-protected area, where 
 humus and partly decomposed plant material had accumulated. All attempts 
 at artificial reseeding failed. 
 
 Area 24 : This area of 140 acres, burned in 1937, was similar to that just dis- 
 cussed, except that the cover consisted of a more balanced mixture of dwarf 
 interior live oak, wedgeleaf ceanothus, Parry manzanita, and herbs. After 
 burning, the live oak clearly predominated, and soon crowded out most of the 
 herbaceous vegetation. This rapid replacement of the herbaceous cover re- 
 duced the grazing capacity, so that, in 1939, 6 acres per animal unit per month 
 was required. The area was used for spring grazing by cattle. The average slope 
 was 20 per cent. Soil erosion was moderate, and the soil was heavily baked. 
 
 Area 25 : The extensive, hilly, red-soil country, extending from the village 
 of Beegum to Ono, was rather well characterized by this 350-acre area. Before 
 burning, in the fall of 1937, various openings supported grass and other herbs 
 with densities ranging from 10 to 30 per cent, whereas under the heavy brush 
 there was but little herbaceous vegetation of any kind. Two years after burn- 
 ing, chamise and sprouting forms of manzanita largely reoccupied the ground, 
 forming a density of about 40 per cent ; but interspersed between the clumps 
 were many broad-leaved herbs and a sprinkling of annual grasses. Hill lotus 
 was perhaps the most valuable forage plant, although some palatable browse 
 was available. Cattle had been grazed upon the area for a short time in the 
 spring. This and similar burns of the locality have afforded satisfactory early- 
 spring goat browsing for three years after burning ; but the lands are regarded 
 by local stockmen as inferior for cattle, and only fair for sheep, except in the 
 draws where a variety of forage was found both before and after burning. 
 About 8 acres of the typical hillside chamise was estimated as required per 
 animal unit per month. The average slope was 30 per cent. Soil erosion in gen- 
 eral was moderate, but was severe on some of the steeper south slopes. Attempts 
 at seeding to cultivated grasses had consistently failed on this area. 
 
 Tehama County. — Area 26 : There was evidence that a lava flow had covered 
 this general region within recent geological time. The area specifically con- 
 
106 University of California — Experiment Station 
 
 cerned consisted of 1,200 acres. The soil was a dark brown, coarse, rocky loam, 
 little more than 3 inches deep, with many large rocks and boulders in evidence. 
 The cover consisted of interior live oak, California scrub oak, species of man- 
 zanita, yerba santa, poison oak, and various annual grasses. After burning 
 there remained a tangle of partially dead woody vegetation, as inspection 
 revealed the following summer. Because of the small amount of forage 
 growth, no stock was placed on the area in 1939 or in 1940. At the time of 
 inspection the main reproduction consisted of California scrub oak sprouts, 
 and of manzanita. The average slope of this area was 12 per cent. There was no 
 abnormal soil loss by erosion, in part perhaps because the rainy season was 
 nearly over when the burning was done, and partly because of the covering of 
 rock fragments and pebbles. Because of inferior soil, the returns from this 
 burn were particularly poor. 
 
 Area 27 : Before burning in 1938, the vegetation on this 30-acre area con- 
 sisted chiefly of blue oak, poison oak, and coffeeberry, with a scattered under- 
 story of foxtail fescue. After burning the area was grazed by sheep. At the 
 time of inspection, sprouts of blue oak, poison oak, and annual grasses occu- 
 pied approximately 30 per cent of the area as a result of the opening up of the 
 brush. The grazing requirement per animal unit per month was 2% acres. 
 The average gradient was 12 per cent. The soil, an alluvial deposit, averaged 
 approximately 1V4 feet in depth. There was little evidence of abnormal ero- 
 sion, probably because the tract is nearly level. Burning temporarily improved 
 the grazing capacity of this area without lowering its productivity. 
 
 Area 28 : A dense stand of canyon live oak and Parry manzanita originally 
 occupied this J ,200-acre area. The burning was done in 1936, following which 
 the area was grazed in the spring period by cattle. In 1939, the area supported 
 a sparse stand of wild oat, soft chess, ripgut grass, and foxtail fescue; but 
 sprouts of canyon live oak and seedlings of Parry manzanita were rapidly 
 reoccupying the ground. The total density of the vegetation was 30 per cent. 
 About 5 acres was required per animal unit per month. The average slope was 
 30 per cent. The clay loam soil, of the Hugo series, varied in depth from 4 
 inches to 2 feet. There were healed-over gullies as evidence of past erosion, but 
 few gullies were actively eroding at the time of inspection ; sheet erosion, too, 
 was light when the area was inspected. 
 
 Amador County. — Area 29 : Before burning, this 260-acre brush field sup- 
 ported a full stand of chamise, greenleaf manzanita, poison oak, and toyon. 
 In 1939, one year after burning, the plant cover, consisting chiefly of toyon 
 and chamise sprouts and seedlings, was sparse, with the extensive areas nearly 
 barren. In 1938, timothy and red clover were sown, but no plants became 
 established from this effort despite the fact that stock had not been placed on 
 the area because of the limited forage growth. The combined density of the 
 browse and herbaceous vegetation amounted to only about 25 per cent cover. 
 The average slope was 15 per cent. The stony, loam soil, of the Auburn series, 
 did not exceed 8 inches in depth, and was conspicuously rocky. Soil loss by 
 sheet and gully erosion was classed as moderate. 
 
 Area 30 : This chamise range of 50 acres was burned in 1935. By 1939, it had 
 been almost completely reinvaded by a dense stand of chamise sprouts and 
 seedlings. The herbaceous vegetation consisted merely of small patches of 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 107 
 
 annual grasses, and most of these were being replaced by the brush. The total 
 density was 50 per cent. No stock had been grazed on the area because of the 
 inferior quality of the forage. The average slope was 20 per cent. The soil, a 
 grayish-black clay loam, did not exceed about 9 inches in depth. There was 
 little evidence of an abnormal rate of erosion. 
 
 Area 31 : A full chamise stand occupied this 200-acre area before burning in 
 1936. At the time of the 1939 inspection, the chamise, interspersed with yerba 
 santa and toyon, had reclaimed the area to the virtual exclusion of herbaceous 
 vegetation, which was represented chiefly by scattered patches of nitgrass, 
 foxtail fescue, and wild oat. For two years after the fire, according to the 
 statement of the owners, this burn carried a few stock, but the forage decline 
 was so rapid that no animals were grazed in 1939. At that time not less than 
 7 acres would have been required per animal unit per month ; the season of use 
 would be short, and the quality of forage inferior. The average slope was 5 per 
 cent. The dark grayish-brown sandy loam showed only slight evidence of 
 abnormal erosion. 
 
 Calaveras County. — Area 32 : Before burning in 1938, the vegetation on 
 this 80-acre area consisted of a dense stand of chamise and a few annual 
 grasses. At the time of inspection, chamise sprouts and seedlings made up 
 approximately one fifth of the total vegetation. The understory cover of fox- 
 tail fescue, nitgrass, wild oat, red brome and chaparral cottonweed, together 
 with the brush sprouts, formed a density of 30 per cent. Cattle grazed the burn 
 lightly in the spring, 4 acres being required for each animal unit per month. 
 The average slope was 15 per cent. The grayish-brown, sandy clay loam varied 
 in depth from 4 to 18 inches. Soil erosion had not exceeded the normal rate 
 after burning. 
 
 Area 33 : Since the fire of 1938, this heavily stocked chamise area of 60 acres 
 had been heavily overgrazed by cattle ; indeed many chamise sprouts had been 
 cropped down to the basal crown. Apparently because of the presence of large 
 accumulations of ash, many spots several feet in diameter were devoid of 
 vegetation. On sites where the fire had been less intense, the vegetation con- 
 sisted of a mere scattering of foxtail fescue, nitgrass, red brome, silver hair- 
 grass, ticklegrass, and wild oat. It may be assumed that if this range had not 
 been so severely overgrazed the 10 per cent density of the herbaceous forage 
 at the time of inspection would surely have been considerably greater. Compu- 
 tations showed that with this inferior forage 6 acres per animal unit per month 
 was required. The average slope was 10 per cent. The light yellowish-gray, 
 gravelly clay loam soil varied in depth from 6 to 24 inches, and showed only 
 light sheet and gully erosion. 
 
 Area 34 : Three years after the fire, chamise had again virtually recaptured 
 all available space. It varied in height from about 6 inches for the seedlings to 
 24 inches for the sprouts. With a ground cover of 40 per cent, composed pri- 
 marily of chamise and annual grasses, 4% acres was required per animal 
 unit per month. Cattle were grazed upon this unit during spring and early 
 summer. The average slope was 25 per cent. The reddish-brown, coarse 
 gravelly clay soil varied from exposed parent rock material to 3 feet in depth, 
 with much rock outcrop. Current soil erosion was classed as moderate ; sheet 
 erosion was conspicuous, and there were several small gullies of recent origin. 
 
108 University of California — Experiment Station 
 
 Summary of Field Observations on Burns. — Of the 34 ranches examined, 
 29 showed rapid reoccupation of the original brush. On 7 of the burns no 
 animals were grazed because of the small amount of good forage produced. 
 Twenty-two of the 34 burns showed only normal or light erosion. It is signifi- 
 cant that the average gradient of these areas was only about 15 per cent. The 
 tabulated records revealed that 30 of the burns had slight or no soil slippage, 
 whereas on 4 areas soil slippage had been severe. Twenty-two burns had little 
 or no gully erosion, whereas on 4 burns, soil erosion of a general nature was 
 severe, and on 8 burns, soil erosion was moderate. The 4 severely eroded areas 
 had an average slope of approximately 44 per cent, whereas those classed as 
 moderately eroded had an average slope of about 32 per cent. Thus the extent 
 of soil erosion of these areas, under grazing use, seemed to show rather strong 
 correlation with steepness of slope, with the amount of ground cover, and with 
 the intensity of grazing. Artificial reseeding had been fully successful on only 
 2 sites, both of these being in the humid coastal region of Humboldt County. 
 On the several other burns where artificial reseeding was tried the results 
 showed only slight and purely temporary success, or they were complete fail- 
 ures from the outset. The estimated acreage required for pasturage of an 
 animal unit was especially high on burns in Lake, Mendocino, Shasta, and 
 Tehama counties ; thus many of these areas had to be classed as inferior for 
 livestock grazing. In Humboldt County, on the other hand, the soils of the 
 areas inspected were productive, and the carrying capacity was high ; hence 
 these burns proved profitable. 
 
 The two factors most favorable to burning were that the animals could graze 
 over most of the area, whereas this was often not possible before burning ; and 
 that more palatable vegetation generally became available after burning. 
 Perhaps the two most adverse factors noted on the controlled burns examined, 
 on the other hand, were the temporary nature of the forage produced and the 
 shortness of the grazing season, which on most of the areas was only a few 
 weeks in the spring, or from spring into the early summer. Only on fairly level 
 sites, and on the more productive soils, was there measurable success from 
 broadcast burning. In general, and particularly on the steeper south and west 
 slopes, heavy or nearly pure stands of chamise, which characteristically occu- 
 pies thin soils, produced the least forage after burning of all brush covers here 
 examined. Many such burns yielded so little forage that no stock was grazed 
 upon them. 
 
 METHODS OF BRUSHLAND IMPROVEMENT 
 
 The studies described in the previous sections have largely been concerned 
 with the effects of the common practice of broadcast burning. In this section 
 are presented a number of current practices employed to maintain the forage 
 on the brushlands when the chaparral stand has been opened up. Some of the 
 methods discussed or proposed involve the use of fire alone ; others are com- 
 binations with mechanical or biological means of controlling the brush. 
 
 Frequent Burning to Keep Cover Open. — Burning chaparral areas at fre- 
 quent intervals is occasionally practiced with a view to decimating the chap- 
 arral sprouts, killing the new crop of seedlings, and removing as many of the 
 old dead stems as possible. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 109 
 
 Where burning at intervals of two to three years is possible, stands of 
 sprouting chaparral tend to give way more or less to grass and weeds. Few 
 such frequently burned areas, however, produce enough herbaceous growth to 
 assure the running of sufficiently hot fires to consume the brush seedlings, the 
 sprouts, and the charred, dead stems. Only on the better sites may a follow-up 
 fire be hot enough, after the two years or so of protection, to destroy most of 
 the incoming chaparral seedlings, and at the same time reduce a portion of the 
 sprout growth. The availability of pasturage elsewhere will determine whether 
 the forage produced after the first fire can be sacrificed solely for improvement 
 purposes. 
 
 Although most of the smaller leafy branches, and much of the accumulated 
 litter, are consumed by the initial fire, a large proportion of the coarser brush 
 stems remain erect and intact five to eight years after burning. Since they are 
 an annoyance to the stock, as well as to the operator, disposal of them is 
 desirable. 
 
 Standing, as these bare stems do, well above the current growth, it is impor- 
 tant that some mechanical means be employed to break them down before 
 reburning. Thus, in late summer or early autumn of the second or third year 
 following the first fire, the stems may be leveled to the ground by dragging a 
 log over the area with a tractor. 28 This method of leveling the stems has been 
 employed successfully by several operators. After the dragging operation, 
 the area is reburned. With ample dry fuel this second fire consumes the old 
 stems and snags, as well as the young brush seedlings, and favors further in- 
 vasions of herbaceous plants. 
 
 The success of the reburning operation depends essentially upon the growth 
 habits of the brush. One hot second fire, or follow-up burn, may transform 
 what was formerly a well-nigh useless nonsprouting brush area to a fairly per- 
 manent cover of herbaceous vegetation. Far less satisfactory results are 
 obtained on areas which support sprouting forms of brush ; few of the plant 
 crowns are killed by the burning, and sprouting continues. Also additional 
 brush seedlings come in after the second burn. Unless the brush seedlings and 
 sprouts are kept closely browsed, as they seldom are, the chaparral soon 
 recaptures the area at the expense of the herbaceous plants. Since livestock, 
 particularly cattle, mostly consume only the herbs, the brush is favored in 
 survival. 
 
 Overstocking with cattle or sheep to kill sprouts and brush seedlings is 
 sometimes resorted to for two to three years after burning. This practice, how- 
 ever, except where goats are used, has generally resulted in failure to destroy 
 the brush, except in nonsprouting cover, where heavy, continuous browsing 
 may eventually destroy all but the less palatable chaparral seedlings. Unfor- 
 tunately very heavy grazing by cattle and sheep also results in destruction or 
 thinning of the grass, and often favors invasions of such unpalatable plants as 
 tarweed, the buckwheats, Napa star thistle, and turkey-mullein, over that of 
 the choicer forage species. Moreover, losses of stock from consuming poisonous 
 plants, and damage of soil erosion on the steeper slopes, have been noted. 
 
 28 The cost of leveling charred stems is about 50 cents per acre, where a stockman uses his 
 own equipment to drag a log about 16 feet long by 14 to 16 inches in diameter across the 
 burned areas. 
 
110 University of California — Experiment Station 
 
 Rotation Burning. — Ranch owners who favor somewhat regular burning of 
 their brush fields, and whose properties are so situated that the chaparral 
 lands constitute an integral and essential part of the grazing unit, should have 
 a burn as young as possible available each year in order to procure maximum 
 forage from those areas. On productive lands burning of the entire brushfield 
 in one fire might provide more seasonal forage than could be utilized to 
 advantage. Following this period of lush growth there would be several sea- 
 sons of limited growth of forage while the brush is reclaiming the land, and 
 while it is too diminutive to reburn. 
 
 In order to derive the greatest sustained benefit and return from burning 
 brush areas, one may adopt a rotation plan of broadcast burning. Provided 
 water for the stock is available, the brush field should be divided into units 
 which may be systematically burned at intervals of eight to ten years. Such a 
 plan is illustrated below, in which the area, growing the sprouting kinds of 
 brush, is divided into three units, each of which would be burned at intervals 
 of nine years. 
 
 Year Unit 1 Unit 2 Unit 3 
 
 1943 Burned Not burned Not burned 
 
 1944 Not burned Not burned Not burned 
 
 1945 Not burned Not burned Not burned 
 
 1946 Not burned Burned Not burned 
 
 1947 Not burned Not burned Not burned 
 
 1948 Not burned Not burned Not burned 
 
 1949 Not burned Not burned Burned 
 
 1950 Not burned Not burned Not burned 
 
 1951 Not burned Not burned Not burned 
 
 1952 Burned Not burned Not burned 
 
 1953 Not burned Not burned Not burned 
 
 Unit 1, or approximately one third of the brush field, is selected for broad- 
 cast burning in 1943. This portion must be relied upon to furnish most of the 
 forage of the brush area during 1944, 1945, and 1946. In 1946, unit 2 is burned. 
 Thus, in 1947, unit 1 should furnish some pasturage, whereas unit 2 should 
 produce a relatively large amount of forage. In 1949 unit 3 is burned ; hence in 
 1950 unit 2 should still be producing some forage, although the major pasture 
 use would be expected from unit 3. In 1952 the burning rotation starts all over 
 again, since in the intervening nine years enough brush growth will have 
 taken place on unit 1 for effective reburning. 
 
 A plan of rotation burning, such as proposed, must not be construed as a 
 method of ridding an area of sprouting brush and permanently replacing it 
 with grass. Within about five years after burning, the brush may have reoc- 
 cupied the area even more densely than before burning. In the interim, how- 
 ever, there should have been an increase of palatable forage for a period of 
 two to four years, perhaps without excessive cost to the operator. 
 
 Cost of Broadcast Burning. — Actual cash outlay by landowners for broad- 
 cast burning of brush fields is usually small, partly because they use rancher 
 cooperative help during the actual burning. Most of the labor involved con- 
 sists of building fire breaks prior to burning. From 7 to 12 men, according to 
 the control necessary, are generally adequate to burn an area of about 100 
 acres. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 111 
 
 Prior to burning, the cost for construction of the necessary fire breaks, as 
 directed by the state forest ranger, varies from about 25 cents to 85 cents an 
 acre — the cost depending on topography, natural barriers, character of bor- 
 derland areas, season, kind of brush, and the type of adjoining lands that must 
 be protected from fire. The range in cost of broadcast burning is illustrated by 
 two cases given below; the data were furnished by the State Division of For- 
 estry. 
 
 Case 1. White Cottage Eanch, Napa County; cost of burning standing chaparral on a 
 100-acre area: 
 
 Per 100 acres Per acre 
 Construction of fire lines, 2 men at 35 cents an hour, 
 
 4 days of 8 hours each $ 22.40 
 
 Actual control time, 5 men at 35 cents an hour, 
 
 3 days of 12 hours each 63.00 
 
 Cost per acre to stockman $0.85 
 
 State Fire Control man, 5 days at $130 per month 21.65 
 
 Travel expense and car 20.00 
 
 Cost per acre ot state 0.42 
 
 $ 0.00 
 
 
 37.80 
 
 
 
 $0.25 
 
 21.65 
 
 
 20.00 
 
 
 
 0.28 
 
 Total costs $127.05 $1.27 
 
 Case 2. Linser Eanch, Napa County; cost of burning standing chaparral on a 150-acre area: 
 
 Per 150 acres Per acre 
 
 Construction of fire lines 
 
 Actual control time, 3 men at 35 cents an hour, 
 
 3 days of 12 hours each 
 
 Cost per acre to stockman 
 
 State Fire Control man, 5 days at $130 per month 
 
 Travel expense and car 
 
 Cost per acre to state 
 
 Total costs $79.45 $0.53 
 
 The rancher must pay all expenses incurred in burning, except those of the 
 special fire-control man, who is paid by the state. 
 
 Although the Linser area was 50 per cent larger than the other, the total 
 cost to the stockman of burning the area was less than half that for the smaller 
 one. This difference in cost was due to variations in natural conditions, such as 
 character of cover, topography, burning weather, natural barriers, and season 
 of the year. The last two factors directly affected the labor expenditure for 
 firebreaks and for fire control. 
 
 The two examples, as indicated, show that the cost depends largely upon 
 local conditions and that each proposed burn presents specific problems. 
 Hence no set rule or formula to compute cost can be set down which will cover 
 all cases. In the two extreme cases cited, the total average cost of burning 
 would be 90 cents per acre, of which the cost to the owner would be 55 cents 
 per acre. 
 
 REMOVAL OF BRUSH AND TREE GROWTH BY METHODS 
 OTHER THAN BROADCAST BURNING 
 
 Scattered throughout the oak- woodland and chaparral associations are some 
 areas of deep soils which are much above average in productivity. Necessary 
 
112 
 
 University of California — Experiment Station 
 
 administrative restrictions by the state on season and locality of broadcast 
 burning, as well as the risk of such burning, have resulted in the initiation of 
 various fire-safe methods by ranchers for removing the brush from these better 
 lands. Where areas are being cleared for a specific form of cultivation, the 
 initial cost of the removal of the brush is often a relatively small factor as corn- 
 
 Fig. 40. — A, Interior live oak felled in winter and permitted to dry until 
 autumn, when the area was burned. B, Oak stumps on part of area shown in A, 
 in process of removal by further burning. Natural grass cover was more than 
 doubled by this clearing process. Mendocino County. 
 
 pared with the returns subsequently obtained from the land. When perma- 
 nently cleared of the brush and tree growth, such areas become a highly 
 important and continuously reliable asset to the owner. But where brush is 
 removed merely to increase the grazing capacity, and the land is not to be cul- 
 tivated, the cost of clearing is speculative. In any case, however, such cost 
 must not be disproportionate to the value of the land. Clearing methods 
 which are in fairly common use are still in the developmental stage, and are 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 113 
 
 Fig. 41. — Eedwood slope logged off, followed by heavy growth of blue blos- 
 som and redwood sprouts, which must be chopped and burned if area is to be 
 opened up. A, At left is shown the lopped and burned portion, with good grass 
 cover ; at right, a charred tangle of brush is left standing after broadcast burn- 
 ing. B, Area temporarily cleared by chopping and burning, at a cost of about 
 $10 an acre. After burning the lopped, dry brush late in the autumn, the area 
 was heavily seeded to redtop, perennial ryegrass, smilo, orchard grass, and 
 common yellow mustard. A good stand of these plants was procured the first 
 year after seeding, but in the third season the introduced plants had almost 
 completely given way to annual grasses and weeds. Coastal region, Mendocino 
 County. 
 
 based more upon initial cost of removal of the brush than upon long-time 
 returns from the cleared lands. Brush-disposal practices, other than by broad- 
 cast burning, may be grouped under three headings : hand methods, mechan- 
 ical methods, and biological methods. 
 
114 University of California — Experiment Station 
 
 Hand Methods. — Hand methods include lopping, cutting, and girdling of 
 brush and trees. One or more of these operations may be used in conjunction 
 with burning, and are frequently an essential part of the other methods de- 
 scribed later. 
 
 Lopping 29 : This procedure is too expensive for general use on dense chapar- 
 ral lands. Even on oak-woodland areas the cost of lopping is prohibitive, 
 except where the soil is unusually fertile, as oak brush can seldom be cleared 
 for less than about $20.00 per acre (fig. 40). On logged redwood lands, how- 
 ever, the common blue blossom can be lopped at a cost of $7.00 to $10.00 an 
 acre. After lopping, the blue blossom brush is allowed to dry until autumn, 
 when it is burned. The large numbers of young lush seedlings of this species 
 
 Fig. 42. — A, Angora goats in grass-Avoodland cover feeding upon brush of 
 ceanothus, manzanita, and wild rose. The central grassy portion of the area for- 
 merly supported a heavy brush stand, but the goats killed sprouts and seedlings 
 after the bushes were chopped and burned. Tehama County. B, Two-year-old 
 sprouts from a stump of interior live oak. Such sprouts are killed by continuous 
 close browsing by goats. Live-oak sprouts are harder to kill than those of other 
 oaks, and must be cut back if they get beyond reach of the goats. El Dorado 
 County. 
 
 which come in after burning furnish superior browse for sheep and goats for 
 two seasons (fig. 41) . 
 
 In order to kill all the sprouting species of brush, the suckers must be cut 
 back every year or two. As many as three to five cuttings of the sprouts may be 
 required before all the oak, or sprouting chaparral, is killed. Despite the high 
 cost, the subsequent removal of the sprouts is a necessary supplement to other 
 clearing methods. 
 
 Cutting 30 : On many oak-woodland areas, oak trees are cut for commercial 
 fuel in conjunction with clearing of the land. Following the initial cutting, 
 spring sucker growth must be kept down by hand-cutting, or by close browsing 
 by goats (fig. 42). Wooded areas located near a favorable market yield an 
 average of $1.50 to $2.00 per cord stumpage, and usually yield from one to 
 
 29 Lopping, as here used, refers to the removal of brush and small trees by chopping at or 
 near the ground level. 
 
 30 Cutting is a term used to designate the felling of trees of commercial size by sawing 
 or chopping or both. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 115 
 
 three cords of wood per acre. Provided fencing is already adequate, the owner 
 can also make a small profit, or he can at least break even, by pasturing goats 
 on the sprouts. Even the expense of fencing may be offset by the sale of wood. 
 Fuel wood sells for about $8.00 per cord in the field (fig. 43). 
 
 Fig. 43. — A, An area cleared by cutting oak for fuel wood. The land will be 
 covered with oak brush in five or six years if sprouts are not controlled. Goats 
 confined to this area would kill the sprouts in approximately three years. B, 
 Oak woodland in the process of being cleared by cutting and goat-browsing. 
 Note the shade trees in the background. There is a dense stand of California 
 melic grass in the foreground. The ricked wood is to be sold. Sprouting stumps 
 in the background will be killed by goat -browsing. El Dorado County. 
 
 Girdling : This method is used for disposing of trees, or patches of timber 
 on productive lands, with a minimum of expense. Girdling is a relatively inex- 
 pensive method, where it is not the aim to salvage the wood for fuel, posts, or 
 other purposes. Moreover, after girdling the live-oak trees sprout much 
 less than after burning or cutting. 
 
 Digger pine is the only conifer girdled to any extent in the foothills for 
 land-clearing purposes, but occasionally Douglas-fir and redwood are so 
 
116 
 
 University of California — Experiment Station 
 
 treated. Girdling of conifers is done with an ordinary ax, in the spring of the 
 year when the "sap is flowing." The trees usually die in one or two years, and 
 fall to the ground in three or four years after girdling, following which the 
 refuse is burned on the ground. 
 
 The chief advantage of girdling oak trees in a land-clearing operation, as 
 indicated, is to prevent extensive sprouting after removal of the top growth. 
 Not more than about 10 per cent of older live oak trees sprout after being 
 girdled, whereas most of the stumps sprout when the trees are merely felled. 
 In contrast, about 50 per cent of the young live oak trees sprout after girdling, 
 about 5 per cent of the younger white oaks sprout, and it is a rarity to find a 
 mature white oak which sprouts after girdling. Oak trees are girdled by ham- 
 
 
 
 
 ? (:i 'i : 
 
 
 • :,■: ■ '/, ■■■ . ■: 
 
 5 
 
 mtsi *-*i 
 
 ; 
 
 W'*>1 - 
 
 
 , . ■ 
 
 11 j | 
 
 
 
 Ihb 
 
 *; ■ •., *'J 
 
 . , l 
 
 ■** - v, - f 
 
 
 <L,. S : '" : : -. :i - :: " .,;?. : : 
 
 f- $0' 
 
 <>.... 
 
 *?- 
 
 . - 
 
 ' ;> *; ; j^ 
 
 Fig. 44. — A, Girdled blue oak, showing how the bark is broken away from the 
 base of the tree with a poleax. One man can girdle 125 to 150 trees in a single 
 day. B, Digger pine girdled with an ax. Trees thus girdled die and fall to the 
 ground within three years. The wood of Digger pine is valueless; unwanted 
 trees are easily disposed of in this way. Amador County. 
 
 mering away the bark 12 to 16 inches above the base of the tree with the back 
 of a poleax (fig. 44). The initial cost of girdling oak lands of average stand 
 amounts to about $5.00 per acre. With stands of smaller trees, additional work 
 must be undertaken on the area each season until all the sprouts are killed. 
 The individual trees die in one to three years after girdling — most of them the 
 first year — and topple over in four to six years. Since the wood of fallen trees 
 is hard and tough, burning during low-fire-risk periods is recommended for 
 their disposal. If fires are run over the area about three times at two-year 
 intervals after girdling, many of the oak sprouts are killed, and the period 
 required for the trees to fall to the ground is shortened. 
 
 Mechanical Methods. — Mechanical methods are those concerned with re- 
 moval of brush by machinery. Two rather widely adopted brush-removing 
 practices, both involving the use of a caterpillar tractor, are outlined below. 
 
 Bulldozing : This method involves the use of a bulldozing blade fastened to 
 a caterpillar. It can be used on lands up to about 30 per cent gradient. Nearly 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 117 
 
 all kinds of brush, as well as oak and pine trees up to about 8 inches in diame- 
 ter, can be removed. The initial cost of clearing chamise lands in this way, in 
 normal times, averages around $3.50 per acre; for bulldozing out heavy stands 
 of oak brush and trees the cost is much higher. The work is usually done in the 
 
 Fig. 45. — A, Heavy chamise stand bulldozed and plowed after seed of the 
 brush had germinated. B, Area treated two years previously as indicated in A. 
 The volunteer cover consists of annual grasses and weeds of fair to good forage 
 value. Amador County. 
 
 spring when the subsoil is moist enough to afford minimum root resistance, but 
 when the ground has become dry enough to give traction for the caterpillar. 
 The method has sometimes had a disturbing influence upon the soil on hilly 
 land, resulting in erosion. In later operations the soil is plowed or disked; ero- 
 sion loss may be prevented by plowing hilly lands along the contour. The 
 
118 University of California — Experiment Station 
 
 contour furrows also kill the older remaining brush plants by exposing the 
 root crowns, and also kill many chaparral seedlings (fig. 45). The additional 
 cost of plowing and removal of suckers is about $3.00 an acre, making the total 
 cost about $6.50 per acre. Occasionally there is some sprouting after bulldoz- 
 ing where the stand of brush is heavy, or where the root crowDs are not 
 removed. A limited amount of hand work with an adz will usually eliminate 
 such young sprouts and root suckers. 
 
 Caterpillar-cable method : This less popular method involves the use of 50 
 to 75 feet of heavy cable, fastened to the tractor as a dragline. The free end of 
 the cable is hooked around a clump of brush, and as tension is applied by the 
 forward movement of the tractor, the brush is pulled out by the roots. This 
 method works well in heavy stands of manzanita, but it is usually more expen- 
 sive than bulldozing. Where larger brush is eradicated by this method, the 
 disturbing influence upon the soil is sometimes severe since the earth is deeply 
 gouged as the caterpillar gains traction. 
 
 In the mechanical methods described, the brush is windrowed or dragged 
 into large piles, allowed to dry, and burned after the serious fire season has 
 passed. 
 
 Biological Methods. — Brush clearing by so-called "biological" methods is 
 essentially concerned with destruction of sprouts and woody seedlings after 
 the cover has been lopped, chopped, or burned broadcast. Goats, and to a less 
 extent deer, are the most effective animals used in killing interior live oak, 
 which is common to the extensive woodland-grass area, and certain other 
 chaparral plants. 
 
 Clearing by goats : When sprouts of interior live oak begin to appear, after 
 cutting or burning, goats are placed on the area, the aim being to keep the 
 young shoots and seedlings browsed down. The browsing must be heavy 
 enough to prevent the sprouts from getting out of reach of the animals. Thus 
 the goats are restricted to relatively small areas, usually by fencing. In order 
 to destroy the oak sprouts, some areas require three goats to the acre the 
 year round, and others as low as one goat on 2 acres, for the first two years 
 after cutting. In the third and subsequent seasons larger acreage per animal 
 is required to maintain the goats in fairly good condition, for the vegetation 
 gradually changes from a browse cover to one of grass. The lands must be 
 continuously stocked to the full capacity of the browse, usually for three to 
 five years ; otherwise only a portion of the sprouting growth will be destroyed. 
 
 If the area to be cleared is not fenced to assure adequate browsing by the 
 goats, or if the animals are not properly cared for, the owner may receive no 
 direct profit from them, yet he will gain through destruction of some of the 
 brush. True, the construction of goat-proof fence is costly. If the cost of fenc- 
 ing is not taken into consideration, $0.75 to $2.00 profit a year on each nanny, 
 while clearing the brush, may be realized. Profit from goats, however, is likely 
 to be low when the time required to take care of the kids is considered ; a kid 
 crop entails about three times more work than does a lamb crop. The invest- 
 ment in goats, in the necessary permanent improvements, and in fences, makes 
 for fairly high initial clearing costs by this method, but records show it has 
 paid well on fairly productive areas. 
 
 Goats are partial to young sprouts of interior live oak, and to wedgeleaf 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 119 
 
 ceanothus, which commonly grow in combination in the woodland-grass asso- 
 ciation ; they prefer this browse to that of many other chaparral species. They 
 will, however, feed upon young sprouts of many other brush species, such as 
 chamise, lemmon ceanothus, poison oak, and manzanita, where there is limited 
 live oak ; but these plants are not browsed close enough to be killed unless the 
 goats are starved to feed upon them. 
 
 Clearing by deer : Young chaparral sprouts and seedlings on small burned 
 areas may be so heavily browsed by deer as to be destroyed in two or three 
 seasons. The burns must be small enough to encourage complete utilization of 
 all the new woody growth. Under such close cropping, the leaves of chamise, 
 one of the most vigorously sprouting forms, grow stemless over the root crown 
 after the first year, and may be completely eliminated by the end of the second 
 year. Larger burns, which do not compel such concentrated deer browsing, 
 fail to result in clearings. The experience of one Lake County stockman indi- 
 cated that burns of about 6 acres were the maximum which could be cleared 
 by deer on his ranch ; 4 to 5 acres were more likely to be cleared. The density of 
 deer populations chiefly determines the maximum practical size of clearing in 
 different regions. Only on choice sites may heavy grass stands be expected to 
 replace the brush killed out by the deer. 
 
 The higher moisture content of repeatedly browsed and regenerating 
 sprouts and seedlings on burns, associated with relatively low crude-fiber con- 
 tent, and high crude-protein levels, as revealed by chemical analysis, might 
 seem to be reasons why deer browse these small burns so much more closely 
 than the contiguous unburned chaparral. 
 
 COMPARISON OF BRUSH-REMOVAL METHODS 
 
 The advisability of using any one specific brush-clearing method is limited 
 by so many different factors and conditions that only general recommenda- 
 tions can be made. On steep slopes, total removal of brush by any method, fol- 
 lowed by close grazing or cultivation, is usually not advisable because of the 
 danger of excessive erosion. Even on moderately steep slopes, the mechanical 
 methods are not applicable because of difficulties of manipulation, and the 
 danger of soil loss. On the gentle slopes the method most adaptable depends 
 upon a number of other conditions. Cutting or chopping should precede use of 
 the area by goats, but these methods are tedious and fairly expensive ; there- 
 fore they are applicable only on the better lands. 
 
 Heavy broadcast burning, although entailing fire risk, is the least expensive 
 in temporarily opening up the dense stands of sprouting forms of brush ; but 
 the benefits derived seldom exceed a duration of about three to four years. A 
 rotation-burning plan, in which advantage is taken of the temporary influx of 
 herbaceous plants, is useful on areas of sprouting brush, if they are to be used 
 more or less regularly for grazing. On nonsprouting brush areas a two-year 
 burning plan, with little or no grazing during that period, is usually effective 
 in converting the brush cover to grassland. 
 
 Where dense chaparral, or oak browse, occupies soils of high quality, one of 
 the mechanical methods described above is useful, as the brush is immediately 
 and almost completely eliminated, and is followed by strong successions of 
 grasses of fairly high carrying capacity. Care must be taken, however, to fur- 
 
120 University of California — Experiment Station 
 
 row on the contour in order to avoid excessive soil loss by erosion, and to 
 destroy the brush seedlings. 
 
 In localities of heavy deer population, close browsing by these animals is 
 effective in eliminating sprouting brush from small burns. Four to 6 acres 
 has been effectively cleared in a few localities of heavy deer population by 
 this means. 
 
 In the adoption of any method of clearing, it is important to weigh probable 
 returns against the costs. Areas of steep, rocky, and thin soils are wholly un- 
 suitable for mechanical clearing effort, or even for the less expensive broadcast 
 burning. Only highly productive sites should be considered in complete me- 
 chanical clearing. On such sites, broadcast burning, with the hope of only two 
 to four years of increased forage, may be less practical in the long run than 
 the use of one of the more expensive but more permanent mechanical or bio- 
 logical methods described. 
 
 COMPARATIVE GRAZING VALUES OF DIFFERENT 
 BRUSH COVERS 
 
 Relative Forage Values. — Some forms of burned brush stands are clearly 
 superior to others for grazing purposes. Although the relative palatability of 
 the different kinds and combinations of brush merits consideration, other fac- 
 tors may have greater weight in determining the usability of the brush fields. 
 
 The nonsprouting stands of chaparral, restricted in area but generally 
 fairly accessible, with relatively few exceptions have productive soils, and 
 clearly rank first in value among the brush stands, in that they give much 
 the best immediate and permanent results after broadcast burning. The 
 purest and most extensive stands of nonsprouting hard brush occur typically 
 in the lower margins of cutover and burned timberlands, and in the upper 
 stretches of the grassland zone below the timber. Conversion of such lands to a 
 volunteer grass cover is relatively simple and inexpensive. After destruc- 
 tion of the brush by burning, the grass cover soon becomes abundant, and is 
 generally composed of a number of the better annual grasses and broad-leaved 
 herbs, with some scattered perennials. 
 
 The association of interior live oak and blue oak ranks second in forage 
 value after cutting, girdling, or burning. Since this cover is also generally 
 accessible, it is useful for grazing during fall, winter, and spring. The produc- 
 tivity of these lands is fairly high, and may be judged in advance by the 
 luxuriance of the tree growth. After the removal of the trees and the de- 
 struction of the sprouts (p. 71) a permanent, dense, and high-yielding 
 grass cover usually comes in. A carrying capacity of 10 to 20 acres per 
 cow unit for the grazing season of about 8 months is not exceptional on cleared 
 lands in this association. The grazing capacity of many such oak lands has 
 been more than doubled by removal of all but desirable shade trees, and by 
 destruction of the Oak sprouts. 
 
 Chamise and other sprouting chaparral covers, which comprise by far the 
 most extensive part of the chaparral association, are of third rank. Of these 
 lands the more or less pure chamise has an even lower rating than the mixed 
 sprouting chaparral cover. When mixed chaparral stands on moderately fer- 
 tile soils are burned, they usually produce a somewhat larger amount, and 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 121 
 
 greater variety, of herbaceous growth than do those of chamise ; and the assort- 
 ment of browse is more palatable when young. 
 
 The extensive, poorer chaparral and chamise lands leave little in the way of 
 choice. It is the burning of these inferior, sprouting chaparral areas that 
 accounts for much of the adverse opinion concerning the value of brush burn- 
 ing or mechanical clearing. These covers occupy the more rugged lands of the 
 chaparral association, which, generally considered, embraces the most inferior 
 soils found in the foothill regions of the state. Much of the area supporting 
 these covers is relatively inaccessible to livestock. So inferior are many of these 
 lands that removal of the brush by mechanical devices, or even leveling the 
 unburned stems in order to dispose of them by burning a second time, is not 
 justified. 
 
 The soil types of some of the more extensive brush-covered areas appear to 
 influence the results obtained by burning. The Aiken soil series, for example, 
 often referred to as "the red soil," is among the most extensive and the least 
 productive of the soil series in the northern counties of the state. In the inter- 
 mediate to more elevated lands of the foothill belt this soil type is often occu- 
 pied by chamise. A large proportion of these lands, burned or unburned, is of 
 low value for grazing, a fact which many stockmen have come to recognize 
 through experience. The old unburned stands found on these soils contain 
 almost no understory cover, and when burned they frequently produce but 
 little herbaceous vegetation of good forage quality. Two to three years of 
 inferior grazing is about all that can be expected after burning such areas ; 
 they are clearly submarginal for grazing purposes. 
 
 Utilization by Cattle, Sheep, and Goats. — Cattle forage upon a large variety 
 of herbaceous plants common to burned chaparral areas after adapting them- 
 selves to this vegetation. They show strong preference for the young grass, but 
 species of such plants as filaree, lotus, mustard, white hawkweed, and several 
 others are taken throughout the grazing period. Early in the season, common 
 soap-plant, woolly-yarrow, California figwort, Napa star thistle, and a fair 
 number of other such broad-leaved herbs are closely cropped, these being taken 
 with a small proportion of the young chamise, sprouting manzanitas, and 
 other such chaparral sprouts. Later in the season, young shoots of California 
 buckeye, ceanothus, interior live oak, California scrub oak, poison oak, and 
 western mountain-mahogany are also browsed. On unburned chaparral lands 
 cattle are inclined to feed upon the browse at an earlier date than on the 
 burned areas, presumably because of the lesser abundance of understory herbs. 
 
 Sheep feed upon a combined diet of grasses, broad-leaved herbs, and browse, 
 throughout the grazing season, and thus utilize more fully the vegetation of 
 burned chaparral areas than do cattle. Also sheep seem to have less seasonal 
 preference for forage and browse than do cattle; they feed upon a large 
 variety of early-season chaparral sprouts and seedlings, including chamise 
 and broad-leaved herbs. In some instances sheep have been observed to main- 
 tain themselves in good condition on burned areas where cattle have been 
 restless and in poor condition because of unsuitable forage. And where many 
 coarse stems are left on an area after burning, sheep seem to make better use 
 of the forage than do cattle. 
 
 Although goats browse extensively upon chamise and sprouting chaparral 
 
122 University of California — Experiment Station 
 
 lands after broadcast burning, in only a few instances do these animals 
 destroy the brush, except around small camp sites or on holding grounds. Sev- 
 eral factors account for the failure to destroy the brush, foremost of which is 
 the fact that owners of large bands of goats must keep the animals in good 
 condition if satisfactory returns are to be obtained. For this reason, the goats 
 are not held overlong on burned areas, but are moved to fresh areas before the 
 browse of any one portion is closely cropped. Moreover, burned chaparral 
 lands are suitable for goats, as they are for other kinds of livestock, only dur- 
 ing the period when the sprouts are young and succulent. One peculiarity in 
 the utilization of the sprouts of hard brush by goats is that most of this vegeta- 
 tion becomes unpalatable to these animals at a specific growth stage. In feed- 
 ing on chamise, for example, the goats strip the young sprouts of their leaf 
 fascicles with some eagerness for several weeks in the spring. As this vegeta- 
 tion hardens slightly, in April in Tehama and Shasta counties, an entire band 
 of goats, within a day or two, completely loses appetite for this browse, and 
 tends to travel extensively, apparently in search of other feed. The herd must 
 then be removed to a different form of vegetation or the gains, or good condi- 
 tion previously attained on the area, will be lost. In about 2 weeks after 
 removal of the goats, the browsed chamise bushes are ref oliated, thus recuper- 
 ating their vigor. This sudden lack of appetite for chamise also applies to 
 several other species of chaparral, but not to the live oak and white oak cover, 
 which is browsed with gusto throughout the season. 
 
 Poisonous Plants. — The marked influx of such poisonous plants as larkspur 
 and death camas, following burning in some areas, is important in considering 
 forage values on burns. The danger is so great in some localities that stockmen 
 decline to place their stock, especially sheep, on brush areas until the second or 
 third year after burning. Experience has taught them that this delayed use is 
 necessary to avoid heavy losses from poisoning. 
 
 Obviously, not all chaparral areas contain large numbers of poisonous 
 plants. Because of early maturity of the forage, and its marked inferiority at 
 later growth stages, however, there is a tendency to place the animals on 
 burned brush areas too early in the season. Some species of poisonous plants 
 are known to be several times more toxic in the very early growth stage than 
 later. Some such species are among the earliest to appear in the spring ; there- 
 fore, in the absence of ample, wholesome, early forage, toxic or lethal amounts 
 of poisonous plants may be taken. Larkspur, for example, has been reported 
 to be three to ten times more toxic to cattle in the early leaf stage than when in 
 flower, and as much as sixteen times more toxic than when in the late fruit 
 stage (89). Some cattlemen have wisely used the flowering stage of larkspur 
 as a guide in determining when to admit cattle on burned areas. Accordingly, 
 some burned brush areas afford safe and satisfactory pasturage for only about 
 4 to 6 weeks in the late spring. Even so short a period of use, however, is of 
 value under many circumstances, for the brushland pasturage may tide over a 
 period until better forage is available elsewhere. 
 
 Not only are the early weeks of the spring period dangerous in some local- 
 ities, but losses to sheep from feeding upon death camas may be somewhat 
 serious throughout the season. Other poisonous plants may cause losses to 
 cattle. In one authenticated instance an operator in Shasta County placed 
 
Bul. 085] Plant Succession on Burned Chaparral Lands 
 
 123 
 
 1,000 cattle of various ages, and of both sexes, on a chamise area burned the 
 previous fall. Fifty -three of these animals, including both old and young indi- 
 viduals, died mainly from larkspur poisoning during early spring, and many 
 
 Fig. 46. — A, Foothill death camas; B, Meadow death camas; C, Fremont 
 death camas ; D, small-meadow death camas. 
 
 others recovered from poison sickness. This operator placed no cattle on the 
 area in the following seasons. 
 
 Although the toxicity of some poisonous plants, such as death camas, varies 
 little with the season, seasonal change differs between species {65, 66) . Differ- 
 ent parts of the plant also vary in toxicity. 
 
124 University of California — Experiment Station 
 
 Death camas is widespread in distribution ; one or more species may appear 
 among the earliest plants on burned areas in northern California. The various 
 species of these plants differ markedly in their poisonous properties. The large- 
 flowered forms, such as Fremont death camas (fig. 46), with relatively few 
 flowers on an open stalk, are far less toxic than the small-flowered forms, such 
 as meadow, foothill, and small meadow death camas (fig. 46, a, b, and d). The 
 latter have numerous small flowers, more or less congested into a pyramidal 
 inflorescence at the apex of the stem. The greater part of all losses of sheep 
 from these species in California, and in fact throughout the West, is caused by 
 some form of the meadow death camas. 
 
 Summary of Grazing Values. — Cattle feed only limitedly upon species of 
 chaparral, unless forced by hunger to do so. Usually cattle are not maintained 
 in good condition on brush fields unless there is a fair abundance of grass. 
 Sheep, on the other hand, feed throughout the spring and early summer on a 
 combination of herbs and young chaparral browse. They can often be main- 
 tained in good condition on brush areas which are unsatisfactory for cattle. 
 Goats may be maintained upon sprouting chaparral and chamise during the 
 first two years after burning. They are not nearly so effective in clearing the 
 hard brush cover, however, as they are in destroying sprouts of live oak and 
 associated browse plants, since they will only feed for a part of the spring 
 period upon sprouts of chaparral and chamise. During the season of non- 
 palatability, the brush regains its vigor. In some localities poisonous plants 
 are so abundant on newly burned areas that the increased forage produced 
 may be offset by losses and sickness of the stock. 
 
 EFFECT OF FIRE ON WILDLIFE AND RECREATION 
 Wildlife and, to a lesser extent, recreation sites are of importance in some 
 chaparral areas. Considering the general increase in hunting and outdoor life 
 the public in the future is likely to place more value on these resources. Since 
 fire may greatly influence the animal numbers and recreational land values, 
 the more pertinent data on the effects of brush burning are here presented. 
 Effect on Small Mammals. — Various small mammals suffer temporary popu- 
 lation reduction from the effects of fire because of the nature of the habitats 
 which they occupy. Tree squirrels and other tree and brush dwellers, for ex- 
 ample, are reduced in numbers during large conflagrations because they 
 instinctively dash to their homes for protection; naturally such burned areas 
 are not restocked by these animals until the taller vegetation has been re-estab- 
 lished (63, 84, 85). On the other hand, the destruction by fire of the smaller 
 mammals which dwell on the surface is mostly temporary (except rabbits, 
 whose population remains low for several years), for enough animals escape 
 the fire to provide a breeding nucleus (61, 62). Moreover, the natural succes- 
 sion of annual vegetation following burning usually provides an improved 
 habitat for the rapid breeding of most surface dwellers during the following 
 few years {24). Animals that burrow, such as ground squirrels, gophers, bad- 
 gers, and some species of mice are little harmed directly by fire (43, 85). Thus 
 Horn, 31 working on experimental chaparral burns in northern California, 
 
 31 Unpublished data. E. E. Horn, U. S. Fish and Wildlife Service. University of Cali- 
 fornia, Berkeley. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 125 
 
 found by actual trapping records covering several successive years, that mice 
 and squirrels soon reached higher populations on the burned areas than 
 occurred on adjoining unburned plots. Usually enough food was found to 
 remain on newly burned chaparral lands to support the surviving mammals. 
 The food supply of those rodents that go into summer and autumn dormancy 
 underground, such as various small ground squirrels, are not immediately 
 affected by fire, since they do not emerge until a new crop of vegetation is 
 available. 
 
 Since the larger predators, such as coyote and fox, are relatively mobile, 
 they are little affected by fire unless extensive areas are burned. These crea- 
 tures temporarily move to adjacent food sources of unburned areas, but 
 return when their prey increases. 
 
 The destruction of small mammals has a secondary detrimental effect on 
 carnivorous fur-bearing animals. Many such mammals may escape the direct 
 effect of the fire, but the reduction of their food supply, notably that of the 
 rodents, causes the fur-bearers to starve after large fires. Where the acre- 
 age burned is small these animals migrate to more favorable adjoining 
 habitats (63). 
 
 Effect on Deer Population. — Extensive fires are known to kill many deer, 
 whereas small fires are more often beneficial than harmful. Although only 
 limited data are available to show the effect of extensive chaparral fires on 
 deer, reports indicate that the direct kill due to large fires is often great. After 
 the 1928 Cahuila fire, for example, 300 deer carcasses were found on a rela- 
 tively small area (92) ; and the Anderson Valley fire caused the death of 59 
 deer on 212 acres. 32 Loss of forage for deer is often a serious indirect effect 
 immediately after large fires. California deer are more or less dependent upon 
 the lower chaparral areas as a source of food during the winter, 33 these lands 
 being a limiting factor on some deer ranges. Fires occurring on these critical 
 areas aggravate overgrazing by, and starvation of, deer. 
 
 Evidence indicates, on the other hand, that small fires are generally not 
 harmful to deer. The agility of these animals allows them to leave burns of a 
 few acres without being injured. Furthermore, according to Horn (43), these 
 animals usually return to burned brush areas the following year to feed upon 
 sprouting chaparral and other succulent feed. In northern California the 
 writer noted on many occasions that deer fed extensively upon the young 
 chaparral sprout growth. 
 
 Special chemical studies of browse samples have revealed that the nutri- 
 tional level of chaparral may be somewhat extended by encouraging the 
 growth of crown sprouts, especially where such regeneration growth is kept 
 closely browsed down. Just how this finding may be applied to deer manage- 
 ment plans on intensively used areas in the state is yet to be determined. Re- 
 moval of the top growth on small, relatively level areas, by carefully controlled 
 burning, or in some other way, might, perhaps, be instituted without serious 
 deterioration of the site. 
 
 32 Report to the writer from J. R. JIall, Forest Supervisor, Stanislaus National Forest, 
 California. 
 
 33 Conference of writer with Ivan Sack, in charge of Fish and Game Management, 
 Region 5, U. S. Forest Service, 'San Francisco, California, 1939. 
 
126 University of California — Experiment Station 
 
 The attraction of deer to fresh burns was known to the Indians, who evi- 
 dently burned small areas in order to localize deer for hunting purposes. Small 
 burns are usually intensively browsed because of the adjoining protective 
 cover and the relatively high palatability of the chaparral sprouts. The con- 
 centration of deer on small burned areas is often so acute as to cause excessive 
 killing or poaching, and may necessitate closed seasons and strict supervision 
 {45,109,110). 
 
 The attraction of deer to recently burned areas has received some attention 
 as a means of diverting these animals from agricultural lands. However, a sur- 
 vey of the observations of ranchers who have practiced brush burning showed 
 that deer move to lower elevations at certain periods in the winter regardless 
 of the treatment of the brushlands, and that cereal crops, cultivated pastures, 
 and vines are preferred to chaparral, whether the latter has been burned 
 or not. 
 
 Influence on Bird Life. — Game birds are killed in large numbers by fire, 
 and their nests, cover, and food may be largely destroyed (101, 102). Clarke 
 (17) noted that three or four years elapsed before pre-fire numbers of birds 
 occurred again on the burned areas studied. 
 
 The direct kill of game birds by fire has been severe in parts of California. 
 In 1928, for example, more than 4,000 quail were lost in relatively large, rap- 
 idly moving brush fires in southern California ; and a 12,000-acre fire in Yolo 
 County accounted for the death of approximately 3,000 quail (108). The 
 destruction of food and cover by fires is also noteworthy. Stoddard (102) has 
 shown that burning of grass of the previous years' growth is unfavorable to 
 the building of quail nests. Leopold (61, 62) noted the need of luxuriant vege- 
 tation for nests in the spring, and for food and cover in fall and winter. 
 
 Small, controlled fires, where burning is done at the proper season, appear 
 to be beneficial in some areas, in intensive quail management. Schierbeck (92) 
 noted the dependence of some forms of wildlife on food plants produced on 
 burns. Stoddard (102) observed that carefully controlled, small fires on some 
 areas in the South are favorable to the propagation of the bobwhite. Sumner 
 (104) also concluded that burning of patches of an acre or so of brush is bene- 
 ficial to the quail population in California. The aim should be to spot-burn 
 small areas where suitable food plants may supplement the spring and winter 
 quail feed. The adjacent thickets serve as protective cover. Such restricted 
 burning, however, is expensive since the operation must be carried out under 
 careful planning and absolute control. 
 
 Secondary Effects on Fish. — Consideration of the essentials of a good fish- 
 ing stream, and the reactions imposed upon it by a fire, seems important since 
 a number of fishing streams head in or pass through the chaparral areas. 
 Pearse (73) has mentioned various requisites of a fish stream, some of which 
 may be adversely affected by burning, namely : ample, prolific stream-bank 
 vegetation; range in temperature suitable to the species of fish concerned; 
 sufficient oxygen to support aquatic life, without excess of carbon dioxide; 
 water relatively free of suspended silt, ash, and dissolved impurities; and 
 acidity relations which are not beyond the range of endurance of the fish. 
 
 By destruction of stream-bank vegetation, fire reduces the insect supply 
 normally inhabiting this vegetation; and it also affects indirectly temperature 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 127 
 
 and oxygen relations (76). Rise in temperature reduces the dissolved gases in 
 the water and upsets the oxygen balance. The increased supply of organic 
 matter carried into the streams after a fire, undergoes decomposition and 
 thus reduces the oxygen content of the water (19). Fish are less able to with- 
 stand a decreased oxygen supply in alkaline water (116). A hot, extensive fire 
 on a watershed, followed by rains and heavy runoff, may carry such quantities 
 of ash, sand, and silt into the water as to be injurious or destructive to all 
 species of fish. The damage may be essentially mechanical by scouring the 
 streambeds of food, and by silting the water so heavily that fish are unable to 
 respire normally (85). Although the pH concentration of stream water may 
 be measurably changed by the additions of ash from burns, fish seem to endure 
 wide ranges of pH. The pH factor is apparently not of especial importance 
 unless the change is great, such as may occur in some small streams (73). 
 
 Effect on Recreational Areas. — The chaparral lands are used only limitedly 
 for recreational purposes other than for hunting and fishing. The extent and 
 location of the areas so used, however, emphasize the importance of knowledge 
 of the relation of fire to recreation. Studies indicate that extensive brush fires 
 are injurious to the best interests of the tourist trade. Hunting, fishing, relaxa- 
 tion, and enjoyment of natural beauty are the chief attractions of areas sought 
 by the public. Tourist travel to areas recently blackened by burning is greatly 
 reduced both during fires and for some years thereafter. Tourists not only 
 leave areas in or adjacent to fires, but they bring back stories of the unsatisfac- 
 tory condition of the area, with the results that friends and neighbors spend 
 their vacations elsewhere (99) . 
 
 MAGNITUDE OF THE BRUSH-BURNING PROBLEM 
 
 Earlier in this report several important aspects of brush burning were con- 
 sidered with regard to individual areas. In evaluating brush burning from the 
 viewpoint of management and administration, one question in particular 
 arises, namely : What is the importance and the magnitude of brush burning, 
 or brush control, in the chaparral area of the state as a whole ? 
 
 Figure 1 shows that the chaparral association occurs throughout the foothill 
 areas of the state, extending almost from its northern boundary to its most 
 southerly border. The fact that the chaparral cover occurs in many counties 
 might lead to the conclusion that brush control is of importance over the state 
 as a whole. This viewpoint is not supported by the actual conditions. 
 
 Except for isolated spots in other areas, brush burning is largely confined to 
 certain counties in regions I and IV (fig. 1). Region I, embracing Lake and 
 Mendocino counties on the coastal slope, and Tehama and Shasta counties in 
 the interior of the state, constitutes the principal area in which broadcast burn- 
 ing is rather extensively practiced. The northern portion of region IV, from 
 Yuba to Calaveras County, contains most of the brush-burning or brush- 
 control areas of the Sierra Nevada slopes. In the other three regions the 
 necessity of maintaining the natural plant cover intact on important water- 
 sheds and water storage areas precludes intentional burning of standing 
 brush, except in localized areas. The elaborate federal-, state-, and county- 
 sponsored flood-control programs, now well under way in parts of southern 
 California, have provided a network of fire lanes and roads designed to aid in 
 
128 University of California — Experiment Station 
 
 controlling fires which might otherwise spread over extensive areas. Facilities 
 for speedy re-establishment of the cover on areas accidentally burned have 
 also been provided. So strongly entrenched has become the philosophy of fire 
 prevention with the public, and with officials who are responsible for the ad- 
 ministration of these lands on important watersheds in the southern part of 
 the state, that occasional requests by landowners to burn even small areas of 
 brush are usually refused. 
 
 The areas upon which brush burning is commonly practiced involve no 
 more than about 30 per cent of the state's chaparral land. Even in the northern 
 counties where land clearing or brush control now constitutes an important 
 problem, broadcast burning is localized because of wide diversity of land uses 
 and conditions. Where public interests clearly outweigh the probable values 
 that might accrue to the landowner from burning, public opinion, fire damage, 
 and state and local laws, all compel the landowner to curb his fire program. 
 
 The brush-burning problem is further restricted locally by recognition 
 among some stockmen that the cost of the forage obtained by the burning of 
 some chaparral lands is too high owing to its low yield and poor quality, and 
 to the protective and control measures that must be taken. In some areas, how- 
 ever, families are found who own so little open country that they can make a 
 livelihood only by systematically burning the brush. Sympathy for the efforts 
 of these small operators appears in some instances to have assumed such prom- 
 inence that the general importance of brush burning over the state has been 
 exaggerated. 
 
 The most general desire of accomplishment of a brush-clearance program is 
 that of permanent removal of the brush, rather than merely to burn down the 
 cover and await its recovery. A program of permanently clearing away the 
 brush, the techniques of which have been discussed, should obviously be fo- 
 cused upon the more productive areas. Brush -burning or other brush-clearance 
 expenditures on the submarginal lands of thin, often steep-sloping, inferior 
 soils, whose forage yield at best is low and inferior, should be held to the 
 
 minimum. 
 
 SUMMARY AND CONCLUSIONS 
 
 A study of plant succession on burns in the chaparral and oak-woodland 
 associations of California, initiated in 1925, was conducted on sample plots, 
 and the broader phases verified by surveys of ranches, chiefly in the northern 
 counties of the state. The field work was supplemented by laboratory experi- 
 ments. Special background studies were included to aid in the interpretation 
 of the results. 
 
 Extent, Nature, and Uses of Foothill Lands. — Slightly more than 7 per cent, 
 or about 7,300,000 acres of the foothill area of the state, lying below 5,000 feet 
 in elevation, is occupied by chaparral, the species of which are characterized 
 by simple, small, thick leaves, and extensive root systems. This cover occurs in 
 two strips which run nearly the length of the state. Chaparral is represented 
 by two distinct growth forms — sprouting and nonsprouting species. When cut 
 or burned, the sprouting forms send up vigorous shoots from a swollen base or 
 "crown," and therefore are exceedingly difficult to eradicate by fire. In con- 
 trast, stands of the nonsprouting forms are destroyed by cutting, or by a single 
 fire, although the invading seedlings must be killed later on. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 129 
 
 The chaparral association may be divided into five regions according to dif- 
 ferences in climate, topography, soil, and species of brush. Its distribution 
 seems chiefly to be delimited to areas where the temperature does not exceed 
 100° F for long periods and where the average annual precipitation does not 
 fall below about 10 inches, or go much above 35 inches. The three most im- 
 portant uses of the chaparral lands are those of watershed, grazing, and timber 
 production; additional uses are those of playground sites, hunting and some 
 fishing, summer homesites, and areas cleared for permanent residences. Inter- 
 mingled with, or adjacent to the chaparral association are redwood, Douglas- 
 fir, and ponderosa pine forests of present or future commercial value. 
 
 Brush-Burning Viewpoints and Practices. — Contrary to the claims of some 
 persons, burning of brushlands by the Indians evidently was restricted both 
 because of their sparse population in chaparral regions, and because of their 
 mode of life. Fire, when used, was primarily to facilitate hunting and to aid in 
 gathering food plants. Burning for such purposes apparently was not ex- 
 tensive enough measurably to enlarge or to decrease the present chaparral 
 areas, or to keep extensive stands open by repeated burning. 
 
 Broadcast chaparral burning apparently reached maximum proportions 
 through activities of the white settler, in the eighties. Enforcement of fire laws 
 since then, together with the adoption of the permit system of burning, 
 appears to have resulted in some decrease in accidental and in incendiary 
 burning. 
 
 A survey of experience and opinion among 80 stockmen in northern counties 
 concerning broadcast bursh burning indicated that the forage on burns is fair 
 to good in quality during the spring and early summer, and that the carrying 
 capacity is usually increased for two to four years, according to the quality of 
 the soil. They also felt that soil erosion is not generally a serious problem. 
 Although strongly marked differences of opinion were reported concerning 
 the value of burning in different localities, including most of the familiar 
 claims such as increased water supply, protection from destructive fires, and 
 insect, rodent, and rattlesnake control, 18 of the 20 stockmen who favored 
 burning of all forms of vegetation resided in Shasta and Tehama counties. In 
 contrast, several stockmen of other counties felt that burning of the more 
 severe manzanita and chamise sites was not worth the effort. Several also dis- 
 couraged burning of open-brush stands, maintaining that fire merely in- 
 creased the density of such brush areas. The present study has substantiated 
 this belief. 
 
 Plant -Succession Studies. — Few plants of sprouting forms of chaparral 
 were killed, and few such stands were destroyed or materially thinned out by 
 periodic burning, whereas the much more restricted nonsprouting brush was 
 killed by a single hot fire. Following burning, numerous annual grasses, broad- 
 leaved herbs, and brush seedlings appeared on most chaparral lands. On areas 
 of sprouting chaparral, the herbaceous growth generally attained maximum 
 abundance the first or second year, and then declined to a mere scattered 
 stand by about the fifth year, owing chiefly to the suppressive effect of the 
 numerous brush sprouts. In contrast, on areas of nonsprouting brush, most of 
 the invading herbaceous plants continued to increase in density well after the 
 third year. The yield of forage on productive chaparral-supporting soils, both 
 
130 University of California — Experiment Station 
 
 of sprouting* and nonsprouting brush areas, was temporarily increased by 
 burning, whereas the extensive lands of thin and unproductive soils responded 
 poorly regardless of the growth habit of the brush. 
 
 Invasions of chaparral species into brush-bordered grassland occurred com- 
 monly and with some regularity after burning, and were more extensive and 
 dense on the side of the brushfields leeward to the prevailing winds, where 
 abundant seed had lodged. Extensions of brush into grasslands surrounding 
 old chaparral patches corresponded roughly in pattern to that of such border 
 areas burned over. 
 
 Germination of some chaparral species was stimulated by heat. The seeds of 
 most of the chaparral species studied were more resistant to heat injury than 
 were those of associated species of grasses or broad-leaved herbs. Soil tempera- 
 tures recorded during chaparral fires more often exceeded the endurance of 
 seeds of herbaceous plants, but they were less often lethal to seeds of brush 
 species. The results appear to account for the many seedlings of some chapar- 
 ral species which invade burned areas. On grassy woodland areas, fairly open 
 except for occasional clumps of more or less decadent stands of chamise or 
 other aggressive sprouting forms, a single fire sometimes resulted in heavy 
 invasions of chaparral seedlings, and vigorous sprouts of the old bushes. 
 Special effort should be made to keep fire out of such brush-decadent areas. 
 
 In the interior counties artificial reseeding trials, using 24 species, were 
 mostly failures, owing chiefly to the long, hot, dry summers, and frequently to 
 inferior soils. Along the more humid coastal strip, on the other hand, greater 
 reseeding success was attained, but results of distinct economic value were 
 confined to soils of high productivity. Here Harding grass, perennial rygrass, 
 and to a lesser extent, wild oat, soft chess, and neecllegrass, were among 
 the most successful introductions. Further experimental trials with highly 
 drought-enduring forms, or with early-maturing species which may escape 
 the summer drought, seem justified. 
 
 Seeds of common range grasses sown on typical brushland soils, and treated 
 with fairly heavy dressings of chaparral ash, resulted in slightly lower per- 
 centages of seed germination, lower yield, and, in most species, in slower 
 development of the later growth stages, than in untreated soils. Soil applica- 
 tions of ash, simulating in amounts those of the heavier accumulations on fresh 
 burns, brought about chlorosis and death of common pasture grasses before 
 they reached the fruiting stage. Chemical analysis of plant specimens of corre- 
 sponding development stages grown in ashed and unashed soils, respectively, 
 showed no significant difference in mineral content, or in crude protein. 
 
 Chaparral ash was found to be appreciably higher in calcium than in potas- 
 sium. Only on small units of broadcast-burned areas, where abundant ash had 
 accumulated, were the pH values so high as to hold in check, for one or more 
 seasons, normal invasions of grasses and other herbs. Accordingly, attempts at 
 reseeding those freshly burned areas which are heavily strewn with ash had 
 best be delayed until one or more autumn rains have dissipated toxic ash con- 
 centrations. This conclusion is corroborated by leaching experiments with 
 chaparral ash which showed heavy loss of potassium and carbonates, and lower 
 pH values. In general, light, slow-moving fires, which are not so likely to 
 sterilize the soil, are followed by more rapid and denser succession of forage 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 131 
 
 plants than where the fire moves rapidly, creates great heat, and burns out the 
 entire cover. 
 
 Most of the 34 brush-burned ranch areas inspected in 1938, 29 of which had 
 been burned under state permit, produced more forage than did the dense 
 adjoining unburned brush units. The grazing capacity and the quality of the 
 forage of most of these burns, however, were low ; indeed, some were not pas- 
 tured because of the limited growth of herbaceous vegetation. The dominant 
 grasses were mostly the early-maturing annual forms reported on most of the 
 plots especially established to study plant succession. The most adverse results 
 noted on most of these burns were the poor quality and the temporary nature 
 of the forage, the low grazing capacity, and, on several areas, the occurrence 
 of accelerated erosion, notably on the steeper, heavily grazed slopes. Many of 
 these areas had been reseeded to cultivated forage plants, but except on good 
 soils of the coastal counties, where some reseedings were successful, most such 
 trials failed from the outset, or the stands were so readily replaced by volun- 
 teer vegetation that the effort did not pay. 
 
 In late spring several common herbaceous species and some brush sprouts 
 on newly burned areas were slightly higher in moisture content than on 
 adjoining unburned lands, and they were 6 to 10 days later in reaching ma- 
 turity. This prolongation in succulence tended to enhance the nutritive value 
 of the forage proportionately, and presumably it also accounted for its higher 
 palatability in late spring. Thus the three most outstanding benefits of burn- 
 ing of areas with productive soils were : a slightly prolonged period of some- 
 what higher nutrition of the vegetation as a whole ; a temporarily greater 
 variety and abundance of accessible forage ; and greater accessibility of the 
 area to the animals and to the operator. 
 
 Soil-Fertility Studies. — Except in heavy localized spots of ash on new 
 burns, as pointed out, changes in acidity, or pH, of the soil were so slight as to 
 affect little the seed germination and subsequent plant growth. On the other 
 hand, the nitrate content of the upper soil layer of practically all chaparral 
 covers studied was higher on fresh burns. Although these differences were 
 largely or wholly nullified the second and subsequent years after the fire, the 
 increased supply of nitrate nitrogen would seem largely to account for the 
 characteristic robust growth of the invading plants the first year after burn- 
 ing. The extensive, frequently thin Aiken soil series, referred to as the "red 
 soil," was among the lowest in nitrogen content of those studied, both before 
 and after burning; especially was this noted where chamise dominated the 
 cover. Moreover, these soils were conspicuously low in grazing capacity. 
 
 Broadcast-Burning Techniques. — Burning to control invasions of chapar- 
 ral involves consideration of both sprouting and nonsprouting forms. In 
 sprouting cover, periodic burning was generally found only temporarily to 
 suppress the brush. In nonsprouting chaparral, on the other hand, the brush 
 was destroyed by two judicious burnings, preferably two years apart, the 
 second burn being preceded by leveling the charred stems with a "drag." 
 Attempts to destroy brush seedlings and sprouts by heavily overstocking new 
 burns of any kind were not successful because of abundant invasions of un- 
 palatable weeds, some poisonous plants, heavy packing of the soil, and accel- 
 erated soil erosion, notably on the slopes. 
 
130 University op California — Experiment Station 
 
 of sprouting* and nonsprouting brush areas, was temporarily increased by 
 burning, whereas the extensive lands of thin and unproductive soils responded 
 poorly regardless of the growth habit of the brush. 
 
 Invasions of chaparral species into brush-bordered grassland occurred com- 
 monly and with some regularity after burning, and were more extensive and 
 dense on the side of the brushfields leeward to the prevailing winds, where 
 abundant seed had lodged. Extensions of brush into grasslands surrounding 
 old chaparral patches corresponded roughly in pattern to that of such border 
 areas burned over. 
 
 Germination of some chaparral species was stimulated by heat. The seeds of 
 most of the chaparral species studied were more resistant to heat injury than 
 were those of associated species of grasses or broad-leaved herbs. Soil tempera- 
 tures recorded during chaparral fires more often exceeded the endurance of 
 seeds of herbaceous plants, but they were less often lethal to seeds of brush 
 species. The results appear to account for the many seedlings of some chapar- 
 ral species which invade burned areas. On grassy woodland areas, fairly open 
 except for occasional clumps of more or less decadent stands of chamise or 
 other aggressive sprouting forms, a single fire sometimes resulted in heavy 
 invasions of chaparral seedlings, and vigorous sprouts of the old bushes. 
 Special effort should be made to keep fire out of such brush-decadent areas. 
 
 In the interior counties artificial reseeding trials, using 24 species, were 
 mostly failures, owing chiefly to the long, hot, dry summers, and frequently to 
 inferior soils. Along the more humid coastal strip, on the other hand, greater 
 reseeding success was attained, but results of distinct economic value were 
 confined to soils of high productivity. Here Harding grass, perennial rygrass, 
 and to a lesser extent, wild oat, soft chess, and needlegrass, were among 
 the most successful introductions. Further experimental trials with highly 
 drought-enduring forms, or with early-maturing species which may escape 
 the summer drought, seem justified. 
 
 Seeds of common range grasses sown on typical brushland soils, and treated 
 with fairly heavy dressings of chaparral ash, resulted in slightly lower per- 
 centages of seed germination, lower yield, and, in most species, in slower 
 development of the later growth stages, than in untreated soils. Soil applica- 
 tions of ash, simulating in amounts those of the heavier accumulations on fresh 
 burns, brought about chlorosis and death of common pasture grasses before 
 they reached the fruiting stage. Chemical analysis of plant specimens of corre- 
 sponding development stages grown in ashed and unashed soils, respectively, 
 showed no significant difference in mineral content, or in crude protein. 
 
 Chaparral ash was found to be appreciably higher in calcium than in potas- 
 sium. Only on small units of broadcast-burned areas, where abundant ash had 
 accumulated, were the pH values so high as to hold in check, for one or more 
 seasons, normal invasions of grasses and other herbs, Accordingly, attempts at 
 reseeding those freshly burned areas which are heavily strewn with ash had 
 best be delayed until one or more autumn rains have dissipated toxic ash con- 
 centrations. This conclusion is corroborated by leaching experiments with 
 chaparral ash which showed heavy loss of potassium and carbonates, and lower 
 pH values. In general, light, slow-moving fires, which are not so likely to 
 sterilize the soil, are followed by more rapid and denser succession of forage 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 131 
 
 plants than where the fire moves rapidly, creates great heat, and burns out the 
 entire cover. 
 
 Most of the 34 brush-burned ranch areas inspected in 1938, 29 of which had 
 been burned under state permit, produced more forage than did the dense 
 adjoining unburned brush units. The grazing capacity and the quality of the 
 forage of most of these burns, however, were low ; indeed, some were not pas- 
 tured because of the limited growth of herbaceous vegetation. The dominant 
 grasses were mostly the early-maturing annual forms reported on most of the 
 plots especially established to study plant succession. The most adverse results 
 noted on most of these burns were the poor quality and the temporary nature 
 of the forage, the low grazing capacity, and, on several areas, the occurrence 
 of accelerated erosion, notably on the steeper, heavily grazed slopes. Many of 
 these areas had been reseeded to cultivated forage plants, but except on good 
 soils of the coastal counties, where some reseedings were successful, most such 
 trials failed from the outset, or the stands were so readily replaced by volun- 
 teer vegetation that the effort did not pay. 
 
 In late spring several common herbaceous species and some brush sprouts 
 on newly burned areas were slightly higher in moisture content than on 
 adjoining unburned lands, and they were 6 to 10 days later in reaching ma- 
 turity. This prolongation in succulence tended to enhance the nutritive value 
 of the forage proportionately, and presumably it also accounted for its higher 
 palatability in late spring. Thus the three most outstanding benefits of burn- 
 ing of areas with productive soils were : a slightly prolonged period of some- 
 what higher nutrition of the vegetation as a whole ; a temporarily greater 
 variety and abundance of accessible forage ; and greater accessibility of the 
 area to the animals and to the operator. 
 
 Soil-Fertility Studies. — Except in heavy localized spots of ash on new 
 burns, as pointed out, changes in acidity, or pH, of the soil were so slight as to 
 affect little the seed germination and subsequent plant growth. On the other 
 hand, the nitrate content of the upper soil layer of practically all chaparral 
 covers studied was higher on fresh burns. Although these differences were 
 largely or wholly nullified the second and subsequent years after the fire, the 
 increased supply of nitrate nitrogen would seem largely to account for the 
 characteristic robust growth of the invading plants the first year after burn- 
 ing. The extensive, frequently thin Aiken soil series, referred to as the "red 
 soil," was among the lowest in nitrogen content of those studied, both before 
 and after burning; especially was this noted where chamise dominated the 
 cover. Moreover, these soils were conspicuously low in grazing capacity. 
 
 Broadcast-Burning Techniques. — Burning to control invasions of chapar- 
 ral involves consideration of both sprouting and nonsprouting forms. In 
 sprouting cover, periodic burning was generally found only temporarily to 
 suppress the brush. In nonsprouting chaparral, on the other hand, the brush 
 was destroyed by two judicious burnings, preferably two years apart, the 
 second burn being preceded by leveling the charred stems with a "drag." 
 Attempts to destroy brush seedlings and sprouts by heavily overstocking new 
 burns of any kind were not successful because of abundant invasions of un- 
 palatable weeds, some poisonous plants, heavy packing of the soil, and accel- 
 erated soil erosion, notably on the slopes. 
 
132 University of California — Experiment Station 
 
 Ranchers who favor burning, and who must rely largely upon continuous 
 pasturage from these lands, may well consider the adoption of a plan of 
 "rotation" burning to provide the maximum nutritious forage year after year. 
 The plan proposed provides that the brush field be divided into three operat- 
 ing units, and that the burning be done in sequence at intervals of approxi- 
 mately three years, or on a nine-year rotation basis. 
 
 Under state supervision the cost to the owner of broadcast burning in north- 
 ern California has varied from approximately 25 cents to 85 cents per acre; in 
 addition the cost of supervision by the state has ranged from about 27 cents to 
 40 cents per acre burned. Many local factors influence the cost, such as the 
 intervals between burnings; amount of cooperative help available; topog- 
 raphy ; season ; natural barriers ; kind, age, and amount of chaparral ; prox- 
 imity of the area to neighboring ranchmen who do not desire to burn; and 
 proximity to forested lands. Isolated brush areas cost least to burn. The cost of 
 burning, the speculative benefits to be derived, and the responsibility which 
 the individual stockman must assume for damage to other property, chiefly 
 account for the small acreage of controlled burning done to date under official 
 supervision. 
 
 All factors considered, it is evident that the meager forage yield resulting 
 from burning areas of thin, low-producing soils, especially when remotely or 
 inconveniently located with respect to the main operating ranch unit, usually 
 does not pay. The fact that much of the foothill brushland still remains in 
 public domain, or is in railroad ownership, is indicative of the low value of 
 extensive units of this land. 
 
 Brush Removal by Methods Other Than Fire. — To avoid fire hazard, and to 
 procure more immediate and lasting results, the use of brush-clearing methods, 
 other than that of broadcast burning, is sometimes justified on productive 
 lands. Hand methods such as lopping, chopping, and girdling have been 
 effectively used, but each method is limited in application by the cost. In oak- 
 woodland, the cost of cutting may sometimes be offset by the sale of cordwood. 
 Girdling, which is best done early in the spring, is less expensive, and is used 
 where the wood is not to be salvaged. Removal of brush by a bulldozer has been 
 successfully employed in restricted localities. This method is particularly 
 applicable to sprouting-brush areas on fairly level lands of productive soils. 
 Browsing animals, particularly goats, have been used to advantage to destroy 
 sprouts and brush seedlings which come in after girdling, chopping, lopping, 
 or burning cleanups. Enough goats should be held on the area to keep the 
 sprouts and seedlings browsed down closely. Where deer are plentiful these 
 animals will often destroy the brush sprouts and seedlings on small, localized 
 burns. 
 
 Comparative Grazing Values of Brush Covers. — Based upon grazing capac- 
 ity, quality of forage, and accessibility of the lands, the brushland covers 
 may be classified, in the order named, as follows with regard to their relative 
 values when burned or cleared of brush: nonsprouting chaparral; interior 
 live oak and blue oak ; and chamise, together with other sprouting chaparral 
 cover. The last is by far the most extensive. 
 
 Sheep utilize all types of brushland more completely than cattle, and ex- 
 hibit fewer seasonal preferences in their choice of forage. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 133 
 
 In localized areas, regardless of the kind of brush, poisonous plants some- 
 times come in so extensively after burning that stockmen fear to pasture them 
 until the second or third year after the fire, when the toxic species usually 
 decline in abundance. Where the more poisonous forms of death camas occur, 
 grazing on fresh burns by sheep should be avoided. Likewise, on larkspur- 
 infested burns, cattle should be kept off in the spring until the larkspur plants 
 are in the full bloom stage. 
 
 Effect of Fire on Wildlife and Recreation. — Wildlife and recreation sites 
 are important in various chaparral areas. Fire destroys many small mammals, 
 notably brush and tree dwellers, or they may die later from starvation. De- 
 struction by fire of small surface-dwelling mammals, on the other hand, is 
 mostly temporary ; frequently, because of the increased food supply, mice and 
 squirrels soon increase in numbers in excess of those present before burning. 
 Because of their mobility, coyotes and certain other large predators are little 
 affected by fires of ordinary size. Deer are seldom injured by small fires, but 
 extensive burns have sometimes resulted in their starvation, injury, or death. 
 Small openings may appreciably increase the forage for deer, but larger 
 burns destroy the protective cover and temporarily deplete the food supply. 
 Extensive brush fires are also adverse to bird life, whereas small, judiciously 
 placed spot fires may be beneficial by providing secluded feeding areas adjoin- 
 ing the unburned cover. Unfavorable reactions of streams, which may follow 
 extensive brush fires on surrounding watersheds, may decimate the fish popu- 
 lation. Moreover, destruction of natural vegetation by fire greatly lowers the 
 value of otherwise attractive camp and recreation sites, and tends to divert 
 the normal tourist trade from fire-swept regions for some time thereafter. 
 
 Importance of Brush Burning. — Although the chaparral association occurs 
 over extensive foothill areas of the state, brush burning is resorted to some- 
 what commonly in only about 30 per cent of its entire range, mainly the north- 
 ern counties. The geographical limits of burning are essentially determined 
 by the relatively high value of undisturbed chaparral lands, such as watershed 
 in the southern part of the state, and by federal, state, and local interests in 
 some chaparral lands elsewhere. Even in the northern counties, where burning 
 is most prevalent, the practice is not county-wide. Although some large oper- 
 ators fire the brush, the greater number of habitual burners are those small 
 stockmen who own little open land, and who must rely for their pasturage 
 upon more or less submarginal brushlands in order to get some use of their 
 holdings. They commonly burn the brush wherever a fire will go through it, 
 and they can ill afford to apply tested mechanical methods of brush suppression 
 after burning. Even though some of these frugal operators present a difficult 
 land-use problem, they justify the fullest tolerance and cooperation of public 
 agencies. Such local cases, however, appear to have exaggerated the impor- 
 tance of brush burning in various parts of the state. 
 
 Some of the larger and more experienced ranchers are tending to confine 
 brush burning to flats and gentle slopes of good soils, and to follow up with 
 various brush-suppression measures. Such procedure usually justifies the cost 
 of the operation, and frequently alleviates grazing pressure in spring and 
 early summer on the more valuable lands. On the other hand, the burning of 
 poor brush land does not pay. 
 
134 University of California — Experiment Station 
 
 Extent of Summer and Autumn Pasturage on Burns. — The greatest single 
 need of the livestock industry of the interior northern counties is that of ade- 
 quate nutritious forage during summer and autumn. Those who benefit most 
 from grazing their animals on brushfields utilize the burns in late winter and 
 spring, when the succulent growth is highest in nutritiveness and palatability. 
 As summer approaches and the forage dries and the brush sprouts harden, 
 the animals are moved to succulent and nutritious pasturage elsewhere. In 
 contrast, those operators who are forced to hold their stock on the brushland 
 throughout the summer and fall, must either overstock the area, or else 
 cut down the breeding herd to the minimum. In either case the benefits derived 
 from the early-season grazing are often largely lost because of the extended 
 grazing season on poor feed. 
 
 Provision of the animals with ample nutritious summer and fall forage is 
 seldom accomplished by brushland burning ; with few exceptions, the period 
 of nutritious forage of these areas is extended for only a relatively short time 
 by burning. This critical situation of seasonal forage deficiency thus raises the 
 question whether additional suitable permanent summer and fall pastur- 
 age could not be developed. True, most of the high mountain range is already 
 fully stocked ; but with the active water development of the northern region, 
 and the established grazing values of Ladino clover and supplementary 
 grasses, the possibilities of increased acreage of irrigated summer pastures 
 should not be overlooked. The recognized need of a better balanced yearlong 
 forage supply warrants further consideration of the use of water for estab- 
 lishment of permanent valley summer pastures. It should be emphasized that 
 flats or gently sloping lands whose soils are fairly deep but have a layer of 
 fine-textured, dark topsoil, and which have relatively little rock outcrop or 
 stony surface, termed "good" or "productive" soils, usually produce abundant 
 and satisfactory early-season forage after the brush is burned. Such areas may 
 be further recognized by the presence of various species of luxuriant grasses 
 and other herbs under the less dense chaparral and in the openings. On the 
 other hand, those lands whose soils are coarse, thin, rocky, of serpentine origin, 
 or those with abundant exposed subsoil, bedrock, or stones, termed "poor" or 
 "unproductive" soils, yield little forage after burning. Moreover, burning of 
 old, open stands of sprouting chaparral, with a strong unclerstory of forage, is 
 usually harmful because fire invigorates the decadent brush, induces invasion 
 of chaparral seedlings, and subsequently chokes out the forage. Burning of 
 distinctly steep slopes, notably those with thin soils and facing to the south or 
 west, is generally uneconomical if for no other reason than that the forage 
 yield is small, of inferior quality, and matures early. The energy which the 
 animals must expend in grazing upon such areas may largely or wholly offset 
 the temporary advantages obtained. 
 
 The Trend in Brush-burning Practices. — More rational controlled or "pre- 
 scribed" burning, confined essentially to the better chaparral lands and com- 
 plimented with management methods which will favor maximum succession 
 and stability of the herbaceous vegetation, may be expected to supplant hap- 
 hazard burning as experience accumulates. 
 
 The most successful use of fire requires careful planning in advance. The 
 first step is to make sure that the expected benefits from burning, all economic 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 
 
 135 
 
 factors considered, will more than offset the cost. The second step is to decide 
 on the specific area to be burned, and to establish adequate fire breaks which 
 will protect the units that are not to be burned. A third measure is to burn late 
 in the fall those areas where the fire risk is high, and to start the fire when the 
 wind and air humidity favor control of the fire. And, finally, there must be 
 an experienced crew of adequate size to procure as clean a burn as possible 
 and to forestall complications. 
 
 COMMON AND SCIENTIFIC NAMES OF PLANTS 
 MENTIONED IN THIS BULLETIN 
 
 Trees and Shrubs 
 
 Bigberry manzanita (ArctostaphyTxys 
 
 glauca) 
 Bigleaf maple (Acer macrophyllum) 
 Bigpod ceanothus (Ceanothus megacarpus) 
 Blue blossom (Ceanothus thyrsiflorus) 
 Blue oak (Quercus Douglasii) 
 Brewer oak (Quercus Ga?ryanava.r. 
 
 Breweri) 
 Buckthorn (Ehamnus) 
 California black oak (Quercus Kelloggii) 
 California buckeye (Aesculus calif ornica) 
 California hazelnut (Corylus rostrata var. 
 
 calif ornica) 
 California huckleberry (Vaccinium ovatum) 
 California incense-cedar (Libocedrus 
 
 decurrens) 
 California scrub oak (Quercus dumosa) 
 California snowbell (Styrax calif ornica) 
 California snowdrop bush (Styrax 
 
 officinalis) 
 California wild grape (Vitis calif ornica) 
 Canyon live oak (Quercus chrysolepis) 
 Chamise (Adenostoma fasciculatum) 
 Chaparral-broom (Baccharis pilularis) 
 Chaparral coffeeberry (Ehamnus calif ornica 
 
 var. tomentella) 
 Chaparral pea (Pickeringia montana) 
 Chaparral whitethorn (Ceanothus leuco- 
 
 dermis) 
 Christmasberry (Photinia arbutifolia) 
 Coast live oak (Quercus agrifolia) 
 Coast whitethorn (Ceanothus incanus) 
 Coffeeberry (Rhamnus calif ornica) 
 Common manzanita (Arctostaphylos 
 
 manzanita) 
 Common snowberry (Symphoricarpos albus) 
 Creambush (Holodiscus discolor) 
 Cupleaf ceanothus (Ceanothus perplexans) 
 Deerbrush (Ceanothus integerrimus) 
 Deerweed (Lotus scoparius) 
 Digger pine (Pinus Sabiniana) 
 Douglas-fir (Pseudotsuga taxifolia) 
 Dwarf canyon live oak (Quercus chrysolepis 
 
 var. nana) 
 Dwarf interior live oak (Quercus Wislizewii 
 
 var. frutescens) 
 Eastwood manzanita (Arctostaphylos 
 
 glandulosa) 
 Foothill ash (Fraxinus dipetala) 
 
 Fremont silktassel (Garry a Fremontii) 
 Greenbark ceanothus (Ceanothus spinosus) 
 Greenleaf manzanita (Arctostaphylos 
 
 patula) 
 Gregg ceanothus (Ceanothus Greggii) 
 Hairy ceanothus (Ceanothus oliganthus) 
 Hoary manzanita (Arctostaphylos 
 
 canescens) 
 Hoaryleaf ceanothus (Ceanothus 
 
 crassifolius) 
 Indian manzanita (Arctostaphylos 
 
 mewuklca) 
 Interior live oak (Quercus Wislizenii) 
 Jim brush (Ceanothus sorediatus) 
 Laurel sumac (Ehus laurina) 
 Leather oak (Quercus durata) 
 Lemmon ceanothus (Ceanothus Lemmonii) 
 Mariposa manzanita (Arctostaphylos mari- 
 
 posa) 
 Mission-manzanita (Xylococcus bicolor) 
 Mountain maple (Acer glabrum) 
 Mock orange (Philadelphus Lewisii var. 
 
 calif ornicus) 
 Ninebark (Physocarpus capitatus) 
 Pacific madrone (Arbutus Mensiesii) 
 Parry ceanothus (Ceanothus Parryi) 
 Parry manzanita (Arctostaphylos Parryana) 
 Pecho mountain manzanita (Arctostaphylos 
 
 pechoensis) 
 Pointleaf manzanita (Arctostaphylos pun- 
 gens) 
 Poison oak (Rhus diversiloba) 
 Ponderosa pine (Pinus ponderosa) 
 Eamona bush (Ceanothus tomentosus var. 
 
 olivaceous) 
 Eedberry buckthorn (Rhamnus crocea) 
 Redwood (Sequoia sempervirens) 
 Ribbonwood (Adenostema sparsifolium) 
 Serpentine manzanita (Arctostaphylos 
 
 obispoensis) 
 Shagbark manzanita (Arctostaphylos rudis) 
 Squawbush (Rhus trilobata) 
 Spicebush (Calycanthus occidentalis) 
 Stanford manzanita (Arctostaphylos Stan- 
 
 fordiana) 
 Straggly gooseberry (Ribes divaricatum) 
 Stripeberry manzanita (Arctostaphylos 
 
 pilosula) 
 Sugarbush sumac (Rhus ovata) 
 
136 
 
 University of California — Experiment Station 
 
 Trees and Shrubs — (Continued) 
 
 Sugar pine (Pinus Lambertiana) 
 Tanoak (Lithocarpus densi flora) 
 Thimbleberry (Rubus parviflorus) 
 Thinleaf huckleberry ( Vaccinium mem- 
 
 branaceum) 
 Toyon (Photinia arbutifolia) 
 Valley oak (Quercus lobata) 
 Wartleaf ceanothus (Ceanothus papillosus) 
 Wartystem ceanothus (Ceanothus verruco- 
 sus) 
 Wavyleaf ceanothus (Ceanothus foliosus) 
 Wedgeleaf ceanothus (Ceanothus cuneatus) 
 Western chokecherry (Prunus demissa) 
 Western mountain-mahogany (Cercocarpus 
 betuloides) 
 
 White alder (Alnus rhombifolia) 
 
 White fir (Abies concolor) 
 
 Whiteleaf manzanita (Arctostaphylos vis- 
 
 cida) 
 Wild currant (Ribes) 
 Wild rose (Rosa) 
 Willow (Salix) 
 Woollyleaf ceanothus (Ceanothus tomento- 
 
 sus) 
 Woollyleaf manzanita (Arctostaphylos to- 
 
 mentosa) 
 Yellowbrush (Chrysothamnus lanceolatus) 
 Yerba santa (Eriodictyon calif ornicum) 
 
 Broad-leaved Herbs 
 
 Alkali clover (Tri folium depauperatum) 
 
 Bishop lotus (Lotus strigosus) 
 
 Bluebells (Mertensia) 
 
 Blue-eyed grass (Sisyrinchium bellum) 
 
 Blue gilia (Gilia capitatum) 
 
 Bolander bedstraw (Galium Bolanderi) 
 
 Bolander linanthus (Linanthus Bolanderi) 
 
 Bracken fern (Pteris aquilina) 
 
 Bur-clover (Medicago hispida) 
 
 Bush beardtongue (Penstemon Lemmonii) 
 
 Bush monkey-flower (Diplacus aurantiacus) 
 
 California everlasting (Gnaphalium decur- 
 
 rens var. calif ornicum) 
 California figwort (Scrophularia calif or- 
 
 nica) 
 California filago (Filago calif ornica) 
 California goldenrod (Solidago calif ornica) 
 Cantua spurge (Euphorbia ocellata) 
 Chaparral cottonweed (Epilobium minu- 
 
 tum) 
 Chaparral penstemon (Penstemon hetero- 
 
 phyllus) 
 Coast larkspur (Delphinium calif ornicum) 
 Common madia (Madia elegans) 
 Common mullein (Verbascum Thapsus) 
 Common peppergrass (Lepidium nitidum) 
 Common soap-plant (Chlorogalum pome- 
 
 ridianum) 
 Common sunflower (Helianthus annuus) 
 Common vervain (Verbena prostrata) 
 Common yellow mustard (Brassica 
 
 campestris) 
 Cotton-batting plant (Gnaphalium chilense) 
 Coyote tobacco (Nicotiana attenuata) 
 Crown brodiaea (Brodiaea multiflora) 
 Cut-leaved thelypodium (Thelypodium 
 
 lasiophyllum) 
 Dodder (Cuscuta) 
 Dotseed plantain (Plantago erecta) 
 Dwarf Bridges grassnut (Brodiaea 
 
 Bridgesii) 
 Dove lupine (Lupinus bicolor) 
 English plantain (Plantago lanceolata) 
 Field suncup (Oenothera micrantha) 
 Fineleaf lotus (Lotus subpinnatus) 
 Fireweed (Epilobium angusti folium) 
 
 Foothill death camas (Zigadenus pani- 
 
 culatus) 
 Fremont death camas (Zigadenus Fre- 
 
 montii) 
 Fremont globemallow (Sphaeralcea Fre- 
 
 montii) 
 Gamble weed (Sanicula crassicaulis) 
 Gayophytum (Gayophytum sp.) 
 Gold fern (Gymno gramme triangularis) 
 Gold- wire (Hypericum concinnum) 
 Grassnut (Brodiaea laxa) 
 Greenbrier (Smilax calif ornica) 
 Hill lotus (Lotus humistratus) 
 Horned snapdragon (Antirrhinum 
 
 cornutum) 
 Horseweed (Erigeron canadensis) 
 Indian pink (Silene calif ornica) 
 Indian tobacco (Nicotiana Bigelovi) 
 Knotweed (Polygonum) 
 Leaf y gilia (Gilia gilioides) 
 Little-bill loco (Astragalus Gambelianus) 
 Longleaf filago (Filago gallica) 
 Meadow death camas (Zigadenus venenosus) 
 Menzies larkspur (Delphinium Menziesii) 
 Milk-aster (Stephanomeria virgata) 
 Monkey flower (Mimulus) 
 Moth mullein (Verbascum Blattaria 
 Mountain death camas (Zigadenus elegans) 
 Mountain nievitas (Cryptantha ambigua) 
 Mule-ears (Wyethia) 
 Mustard (Brassica) 
 Nada stickleaf (Mentzelia dispersa) 
 Napa star thistle (Centaur ea melitensis) 
 Narrowleaf soap-plant (Chlorogalum 
 
 angustifolium) 
 Northern rockcress (Arabis retrofracta) 
 Nuttall bedstraw (Galium Nuttallii) 
 Prickly lettuce (Lactuca scariola) 
 Purple nightshade (Solanum Xantii) 
 Eattlesnake weed (Daucus pusillus) 
 Ked clover (Trifolium pratense) 
 Eed larkspur (Delphinium nudicaule) 
 Bedmaids (Calandrinia Menziesii) 
 Red ribbons (ClarTcia concinna) 
 Redstem filaree (Erodium cicutarium) 
 Rush lotus (Lotus junceus var. Biolettii) 
 
Bul. 685] 
 
 Plant Succession on Burned Chaparral Lands 
 
 137 
 
 Broad-leaved Herbs — (Continued) 
 
 Sagebrush (Artemisia) 
 
 St. Johnswort (Hypericum perforatum) 
 
 Seouler St. Johnswort (Hypericum formo- 
 
 sum var. Scouleri) 
 Sedge (Car ex) 
 
 Sheep sorrel (Eumex Acetosella) 
 Silverleaf lupine (Lupinus albifrons) 
 Slender buckwheat (Eriogonum gracile) 
 Slender madia (Madia exigua) 
 Slender popcorn flower (Plagiobothtys 
 
 tenellus) 
 Small-flowered lotus (Lotus micranthus) 
 Small meadow death camas (Zigadenus 
 
 venenosus micranthus) 
 Spanish-clover (Lotus americanus) 
 Star thistle (Centaurea) 
 Sulfur flower (Eriogonum umbellatum) 
 Summer cottonweed (Epilobium panicu- 
 
 latum) 
 Sword fern (Polystichum munitum) 
 Tarweed (Hemizonia) 
 Tobacco mimulus (Mimulus Bolanderi) 
 Tomcat clover (Trifolium tridentatum) 
 Torrey nievitas (Cryptantha Torreyana) 
 Turkey-mullein (Eremocarpus setigerus) 
 
 Valley tassels (Orthocarpus attenuatus) 
 Vetch (Vicia) 
 
 Western lupine (Lupinus leucophyllus) 
 Western thistle (Cirsium Occident ale var. 
 
 Coulteri) 
 Whispering bells (Emmenantha penduli- 
 
 flora) 
 White everlasting (Gnaphalium micro- 
 
 cephalum) 
 White-flowered navarretia (Navarretia 
 
 leucocephala) 
 White forget-me-not (Cryptantha afflnis) 
 White hawkweed (Hieracium albiflorum) 
 Whitestem filaree (Er odium moschatum) 
 White sweetclover (Melilotus alba) 
 Wild buckwheat (Eriogonum) 
 Wild carrot (Daucus Carota) 
 Wild tobacco (Nicotiana) 
 Willow herb (Epilobium) 
 Wool-mat (Psilocarphus tenellus) 
 Woolly-yarrow (Eriophyllum lanatum var. 
 
 achillaeoides) 
 Yellow star thistle (Centaurea solstitialis) 
 Yellow sweetclover (Melilotus indica) 
 
 Grasses 
 
 Barley (Hordeum) 
 Bent-head fescue (Festuca reflexa) 
 Blue wild-rye (Elymus glaucus) 
 Brome grass (Bromus) 
 Bulbous bluegrass (Poa bulbosa) 
 California brome (Bromus carinatus) 
 California melic (Melica calif ornica) 
 Canada bluegrass (Poa compressa) 
 Columbia needlegrass (Stipa columbiana) 
 Crested wheatgrass (Agropyron cristatum) 
 Downy chess (Bromus tectorum) 
 Foxtail fescue (Festuca megalura) 
 Harding grass (Phalaris tuberosa var. 
 
 stenoptera) 
 Italian ryegrass (Lolium multiflorum) 
 Junegrass (Koeleria cristata) 
 Little quaking grass (Briza minor) 
 Kentucky bluegrass (Poa pratensis) 
 Malpais bluegrass (Poa scabrella) 
 Meadow fescue (Festuca elatior) 
 Mouse barley (Hordeum murinum) 
 Mountain brome (Bromus marginatus) 
 Needle-and-thread (Stipa comata) 
 Nevada bluegrass (Poa nevadensis) 
 New Mexican needlegrass (Stipa neomexi- 
 
 cana) 
 Nitgrass (Gastridium ventricosum) 
 Orchard grass (Dactylis glomerata) 
 
 Pacific fescue (Festuca pacifica) 
 
 Perennial ryegrass (Lolium perenne) 
 
 Pinegrass (Calamagrostis rubescens) 
 
 Pineland three-awn (Aristida stricta) 
 
 Needlegrass (Stipa) 
 
 Prairie beardgrass (Andropogon scoparius) 
 
 Purple needlegrass (Stipa pulchra) 
 
 Eat-tail fescue (Festuca Myuros) 
 
 Red brome (Bromus rub ens) 
 
 Redtop (Agrostis alba) 
 
 Ripgut grass (Bromus rigidus) 
 
 Sheep fescue (Festuca ovina) 
 
 Silver hairgrass (Air a caryophyllea) 
 
 Six -weeks fescue (Festuca octo flora) 
 
 Slender hairgrass (Deschampsia elongata) 
 
 Slender oat (Avena barbata) 
 
 Slender wheatgrass (Agropyron pauci- 
 
 florum) 
 Smilo (Oryzopsis miliacea) 
 Smooth brome (Bromus inermis) 
 Soft chess (Bromus mollis) 
 Spanish brome (Bromus madritensis) 
 Ticklegrass (Agrostis hiemalis) 
 Timothy (Plileum pratense) 
 Velvet grass (Holcus lanatus) 
 Western needlegrass (Stipa occidentalis) 
 Wild oat (Avena fatua) 
 Wild-rye (Elymus) 
 
138 University of California — Experiment Station 
 
 ACKNOWLEDGMENTS 
 
 In the course of these investigations the author has had the cooperation of 
 many individuals. To Aida Montier goes the credit for the many plant draw- 
 ings. A. H. Gold, H. G. Reynolds, and Dr. 0. S. Walsh assisted in compiling 
 field data, and they were helpful in many other ways. J. 0. Bridges assisted in 
 a major way with the field surveys pertaining to experiences and attitudes of 
 stockmen concerning burning; and P. S. Pattengale assisted with the field 
 observations of areas burned under state permit. To Mrs. Beryl S. Jesperson 
 thanks are expressed for verifying the plant identifications and the common 
 names of plants mentioned in the bulletin. Dr. A. Gordon was helpful in check- 
 ing chemical analyses of plants collected on burned and unburned areas, and 
 in reading the section on this phase of the study. Dr. A. L. Kroeber, Professor 
 of Anthropology, extended to the writer the privilege of reviewing field notes 
 recorded by his staff on the history of Indian burning ; also he made available 
 the archives of the library of his department, and critically read the section on 
 burning by Indians in California. The preparation of the map showing the dis- 
 tribution of the chaparral association was made possible through the courtesy 
 of A. E. Wieslander of the California Forest and Range Experiment Station, 
 United States Forest Service; and he personally checked the distribution of 
 the areas. M. B. Pratt, State Forester, contributed to the statement on the his- 
 tory of state fire-control policies in relation to burning, and also furnished cost 
 data on burning operations conducted under state control. E. E. Horn, Biolo- 
 gist, United States Fish and Wildlife Service, critically reviewed the section 
 on the effect of fire on wildlife, and amplified some of the points therein from 
 unpublished field records. The following colleagues offered constructive sug- 
 gestions on the paper as a whole : Dr. G. B. Bodman, Professor of Soil Physics ; 
 Dr. W. P. Kelley, Professor of Soil Chemistry ; Dr. J. Kittredge, Professor of 
 Forestry; Dr. D. Weeks, Associate Professor of Agricultural Economics; and 
 W. W. Weir, Drainage Engineer. Special acknowledgment is due to H. E. 
 Malmsten for valuable cooperation in the earlier pursuit of the brush-burning 
 project. Mr. Malmsten was active in the field studies for several years, and 
 contributed particularly to certain phases of the plant-succession investigation. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 139 
 
 LITERATURE CITED 
 
 1. Aldous, A. E. 
 
 1934. Effect of burning on Kansas bluestem pastures. Kansas Agr. Exp. Sta. Tech. 
 Bul. 38:1-65. 
 
 2. Alway, F. J., and C. O. Eost. 
 
 1927. Effect of forest fires upon the composition and productivity of the soil. First 
 Internatl. Cong. Soil Sci. Proc. and Papers [Washington, D. C, 1927], 
 3:546-76. 
 
 3. Andrews, H. J., and E. W. Cowlin. 
 
 1940. Forest resources of the Douglas-fir region. U. S. Dept. Agr. Misc. Pub. 389. 
 
 4. Association op Official Agricultural Chemists. 
 
 1935. Official and tentative methods of analysis of the Association of Agricultural 
 Chemists. 4th ed. 710 p. Assoc. Off. Agr. Chem., Washington, D.C. 
 
 5. Bancroft, H.H. 
 
 1886. The native races of the Pacific States — wild tribes. Works of H. H. Bancroft, 
 vol. 1. 797 p. The History Company, San Francisco, Calif. 
 
 6. Barnette, E. M., and J. B. Hester. 
 
 1930. Effect of burning upon the accumulation of organic matter in forest soil. Soil 
 Sci. 29:281-84. 
 
 7. Barrett, S. A., and E. W. Gifford. 
 
 1933. Miwok material culture. Public Museum of the City of Milwaukee Bul. 2: 
 117-376. 
 
 8. Barton, L. V. 
 
 1934. Dormancy in Tilia seeds. Boyce Thompson Inst. Contrib. 6:69-89. 
 
 9. Baxley, H. W. 
 
 1865. What I saw on the west coast of South and North America. 632 p. D. Appleton 
 and Company, New York, N. Y. 
 
 10. Beals, E. 
 
 1933. Ethnology of the Niseman. California Univ. Pubs. Amer. Archaeol. and Ethnol. 
 31(6):335-414. 
 
 11. Beckwith, E. G. 
 
 1854. Eeport of explorations for a route for the Pacific Eailroad on the line of the 
 forty-first parallel. U. S. War Dept. Eeports of Explorations and Surveys 
 1853-56. Vol. 2. 114 p. 
 
 12. Belcher, E. 
 
 1843. Narrative of a voyage around the world, performed in H. M. S. "Sulphur," 
 during the years 1836-1842. Vol. 2. 460 p. Henry Colburn, London. 
 
 13. Benedict, M. A. 
 
 1930. Twenty-one years of fire protection in the national forests of California. Jour. 
 Forestry 28:707-10. 
 
 14. California Department of Natural Eesources. 
 
 1941. Forest and fire laws, 1937. 169 p. California Dept. Nat. Eesources, Div. For- 
 estry, Sacramento, Calif. 
 
 15. California State Board of Forestry. 
 
 1926. Forest and fire laws and regulations, 1925-27. 71 p. California State Board of 
 Forestry, Sacramento, Calif. 
 
 16. Clar, C. E. 
 
 1939. California second-growth pine. 16 p. State Board of Forestry, Sacramento, 
 Calif. 
 
 17. Clarke, W. T. 
 
 1922. Agriculture in cutover redwood lands. California Agr. Exp. Sta. Bul. 350: 
 167-86. 
 
 18. Clements, F. E. 
 
 1920. Plant indicators. Carnegie Inst. Washington Pub. 353:1-388. 
 
 19. Coker, E. E. 
 
 1915. Water conservation, fisheries, and food supply. Pop. Sci. Monthly 87:90-99. 
 
140 University of California — Experiment Station 
 
 20. Cooper, W. S. 
 
 1922. The broad-sclerophyll vegetation of California. Carnegie Inst. Wash. Bui. 
 319 : 1-124. 
 
 21. Costanso, M. 
 
 1911. The Portola expedition of 1769-1770. Diary of Miguel Costanso,- edited by 
 Frederick Teggart. Acad. Pacific Coast History Pub. 2(4) : 1-167. 
 
 22. Crespi, J. 
 
 1927. Diary of the Portola expedition, 1769-1770. 402 p. Univ. California Press, 
 Berkeley, Calif. 
 
 23. Crocker, W. M., and G. T. Harrington. 
 
 1918. Catalase and oxidase content of seeds in relation to their dormancy, age, via- 
 bility, and respiration. Jour. Agr. Ees. 15:137-74. 
 
 24. Ellenwood, F. A. 
 
 1939. Common-sense conservation. Testimony before joint committee on forestry of 
 the United States Congress. 21 p. United Conservation League, Vina, Calif. 
 
 25. Fabricius, L. 
 
 1929. Forstliche Versuche. Die Einwirkung von Waldbrandasche auf Samenkeimung 
 und erste Pflanzenentwicklung. Forstwiss. Centbl. 51(8) : 269-76. 
 
 26. Forsling, C. L. 
 
 1931. A study of the influence of herbaceous plant cover on surface runoff and soil 
 erosion in relation to grazing on the Wasatch plateau in Utah. U. S. Dept. Agr. 
 Tech. Bui. 220: 1-71. 
 
 27. Gibbs, G. 
 
 1860. Journal of the expedition of Colonel Eedick McKee through northwestern 
 California in 1851. Henry E. Schoolcraft, Archives of Aboriginal Knowledge 
 3:99-177. 
 
 28. GORDON-CUMMING, F. C. 
 
 1883. Wild tribes of the Sierras. Natl. Eev. 2(8) :412-21. 
 
 29. Gordon, A., and A. W. Sampson. 
 
 1939. Composition of common California foothill plants as a factor in range man- 
 agement. California Agr. Exp. Sta. Bui. 627 : 1-95. 
 
 30. Greene, S. W. 
 
 1935. Effects of annual grass fires on organic matter and other constituents in virgin 
 longleaf pine soils. Jour. Agr. Ees. 50 : 809-22. 
 
 31. Harris, F. S., and D. W. Pittman. 
 
 1919. Eelative resistance of various crops to alkali. Utah Agr. Exp. Sta. Bui. 168 : 
 3-23. 
 
 32. Hart, G. H., H. E. Guilbert, and H. Goss. 
 
 1932. Seasonal changes in the chemical composition of range forage and their rela- 
 tion to nutrition of animals. California Agr. Exp. Sta. Bui. 543 : 1-62. 
 
 33. Hastings, L. W. 
 
 1845. The emigrants' guide to Oregon and California, containing a description of 
 California. 157 p. George Conclin Co., Cincinnati, Ohio. 
 
 34. Helper, H. E. 
 
 1855. The land of gold. 300 p. Henry Taylor Co., Baltimore, Md. 
 
 35. Hensel, E. L. 
 
 1923. Effect of burning on vegetation of Kansas pastures. Jour. Agr. Ees. 23:631-44. 
 
 36. Hesselman, H. 
 
 1918. Soil nitrification in relation to forest production. Skagsvordsforeningens Ted- 
 skrift. 108 p. Reviewed in: Jour. Forestry 17:69-73. 
 
 37. Heyward, F. 
 
 1936. Soil changes associated with forest fires on the longleaf pine region of the 
 South. Amer. Soil Survey Assoc. Bui. 17:1-190. 
 
 38. Heyward, F. 
 
 1939. Some moisture relations of soils from burned and unburned longleaf pine 
 forests. Soil Sci. 47:313-24. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 141 
 
 39. Hey ward, F., and R. M. Barnette. 
 
 1934. Effect of frequent fires on chemical composition of forest soils in the longleaf 
 pine region. Florida Agr. Exp. Sta. Bul. 265:1-39. 
 
 40. Hinds, R, B. 
 
 1844. The botany of the voyage of the H. M. S. "Sulphur," under command of Cap- 
 tain Sir Edward Belcher during 1836-42. 195 p. Smith, Elder, and Co., London. 
 
 41. Hofmann, J. V., 
 
 1917. Natural reproduction from seed stored in the forest floor. Jour. Agr. Ees. 
 11:1-26. 
 
 42. Hofmann, J. V. 
 
 1924. The natural regeneration of Douglas-fir in the Pacific northwest. U. S. Dept. 
 Agr. Dept. Bul. 1200 : 1-62. 
 
 43. Horn, E. E. 
 
 1938. Some wildlife forest relationships. Third North Amer. Wildlife Conf . Trans, 
 p. 376-80. American Wildlife Institute, Washington, D. C. 
 
 44. Hough, W. 
 
 1926. Fire as an agent in human culture. U. S. Natl. Mus. Bul. 139 : 1-270. 
 
 45. Hunter, J. S., and D. H. Fry. 
 
 1940. Progress report on game kill statistics. California Fish and Game 26:310-33. 
 
 46. Hutchings, J. M. 
 
 1888. In the heart of the Sierras. 496 p. Pacific Press Pub. House, Oakland, Calif. 
 
 47. Ingram, D. C. 
 
 1931. Vegetative changes and grazing use on Douglas-fir cutover land. Jour. Agr. 
 Res. 43:387-417. 
 
 48. Isaac, L. A. 
 
 1930. Seedling survival on burned and unburned areas. Jour. Forestry 28:569-71. 
 
 49. Isaac, L. A. 
 
 1940. Vegetative succession following logging in the Douglas-fir region with special 
 reference to fire. Jour. Forestry 38:716-21. 
 
 50. Isaac, L. A., and H. G. Hopkins. 
 
 1937. The forest soil of the Douglas-fir regions and the changes wrought upon it 
 by logging and slash burning. Ecology 18(2) : 264-79. 
 
 51. Josephson, H. R. 
 
 1941. Factors affecting income from second-growth forests in the western Sierra 
 Nevada. California Agr. Ext. Sta. Bul. 658:1-72. 
 
 52. Kelly, I. T. 
 
 1932. Ethnology of the Surprise Valley Paiute. Calif. University Pubs. Amer. Arch- 
 aeol. and Ethnol. 31(3) : 67-210. 
 
 53. Kern, E. M. 
 
 1865. Indian customs of California. Henry R. Schoolcraft, Archives of Aboriginal 
 Knowledge 5 : 649-50. J. B. Lippincott and Co., Philadelphia, Pa. 
 
 54. Korstian, C. F. 
 
 1927. Factors controlling germination and early survival in oaks. Yale Univ. School 
 Forestry Bul. 19:1-115. 
 
 55. Kotok, E. I. 
 
 1933. Fire, a major ecological factor in the pine region of California. Fifth Sci. 
 Congress Proc. [Victoria and Vancouver, B. C, Canada.] Toronto Univ. Press 
 B7-22: 4017-22. 
 
 56. Kroeber, A. L. 
 
 1925. Handbook of the Indians of California. Bur. Amer. Ethnol. Bul. 78 : 1-995. 
 
 57. Kuttner, T., and H. R. Cohen. 
 
 1927. Micro-colorimetric studies. Jour. Bio. Chem. 75:517-31. 
 
 58. Larsen, J. A. 
 
 1925. Natural reproduction after forest fires in northern Idaho. Jour. Agr. Res. 
 30:177-97. 
 
 59. Larsen, J. A. 
 
 1929. Fires and forest succession in the Bitterroot Mountains of northern Idaho. 
 Ecology 10:67-76. 
 
142 University of California — Experiment Station 
 
 60. Larsen, J. A. 
 
 1935. Natural revegetation on eroded soils in southeastern Ohio. Iowa State Col. 
 Jour. Sci. 9:151-64. 
 
 61. Leopold, A. 
 
 1923. Wild flowers of the forest. The effect of fires on fish and game. Amer. Forestry 
 29:515-19,568. 
 
 62. Leopold, A. 
 
 1931. Report of a game survey on the north central states. 229 p. Democrat Printing 
 Co., Madison, Wis. 
 
 63. Leopold, A. 
 
 1933. Game management. 481 p. Charles Scribner and Sons, New York, N. Y. 
 
 64. LOWDERMILK, W. C. 
 
 1935. Certain aspects of the role of vegetation in erosion control. Iowa State Col. 
 Jour. Sci. 9:123-32. 
 
 65. Marsh, C. D., and A. B. Clawson. 
 
 1922. The death camas species as poisonous plants. U. S. Dept. Agr. Dept. Bui. 
 1012:1-25. 
 
 66. Marsh, C L\, and A. B. Clawson. 
 
 1924. The meadow death camas as a poisonous plant. U. S. Dept. Agr. Dept. Bui. 
 1240:1-14. 
 
 67. Morris, M. S. 
 
 1931. Increasing forage on sagebrush land. Colorado Agr. Col. Bui. 308-A:l-7. 
 
 68. Morrison, F. B. 
 
 1937. Feeds and feeding. The Morrison Pub. Co., Ithaca, N. Y. 
 
 69. Mum, J. 
 
 1907. The mountains of California. 381 p. Century Co., New York, N. Y. 
 
 70. Munns, E. N. 
 
 1923. Report to the legislature on Senate concurrent resolution no. 27, by the Cali- 
 fornia State Board of Forestry. 165 p. Calif. State Printing Office, Sacramento, 
 Calif. 
 
 71. Neal, W. M., and R. B. Becker. 
 
 1933. The composition of the feedstuff's in relation to nutritional anemia in cattle. 
 Jour. Agr. Res. 47:249-55. 
 
 72. Palou, F. F. 
 
 1926. Historical memoirs of new California. Vol. 4. 381 p. Univ. California Press, 
 Berkeley, Calif. 
 
 73. Pearse, A. S. 
 
 1937. Animal ecology. 2d ed. 642 p. McGraw-Hill Book Co., New York, N. Y. 
 
 74. Person, H. L. 
 
 1937. Commercial planting on redwood cutover lands. U. S. Dept. Agr. Cir. 434:1-40. 
 
 75. Plummer, F. G. 
 
 1911. Chaparral. U. S. Forest Serv. Bui. 85 : 1-48. 
 
 76. Powers, E. B. 
 
 1929. Fresh water studies: I. The relative temperature, oxygen content, alkali re- 
 serve, the carbon dioxide tension and pH of the waters of certain mountain 
 streams at different altitudes in the Great Smoky Mountain National Park. 
 Ecology 10:97-111. 
 
 77. Powers, S. 
 
 1873. The California Indians. No. IX. The Yocuts. Overland Monthly 11(2) : 105-16. 
 
 78. Quick, C. R. 
 
 1935. Notes on the germination of ceanothus seeds. Madrono 3(3) : 135-40. 
 
 79. Robbins, W. W. 
 
 1940. Alien plants growing without cultivation in California. California Agr. Exp. 
 Sta. Bui. 637:1-128. 
 
 80. Robinson, L. 
 
 1938. Forestry follies. 38 p. Published by the author, El Toro, California. 
 
 81. Rows, P. B. 
 
 1929. Water supply for rainfall on valley floors. Amer. Soc. Civ. Engin. Proc. 55(2) : 
 2695-2704. 
 
Bul. 685] Plant Succession on Burned Chaparral Lands 143 
 
 82. Eussell, E. J. 
 
 1926. Climates of California. Calif. Univ. Pubs. Geog. 2(4) : 78-84. 
 
 83. Eussell, E. J. 
 
 1931. Dry climates of the United States. Calif. Univ. Pubs. Geog. 5:1. 
 
 84. BUTLEDGE, A. 
 
 1928. Forest fire destroys deer, grouse, bear, and other wild life. U. S. Dept. Agr. 
 Yearbook of Agriculture 1928:315-16. 
 
 85. Eutledge, A. 
 
 1928. Wildlife in a forest fire. Amer. Forests and Forest Life 34:451-53. 
 
 86. Salisbury, E. J. 
 
 1925. Note on the edaphic succession on some dune soils with special reference to the 
 time factor. Jour. Ecol. 13:322-28. 
 
 87. Sampson, A. W. 
 
 1919. Plant succession in relation to range management. U. S. Dept. Agr. Dept. Bul. 
 791:1-76. 
 
 88. Sampson, A. W. 
 
 1923. Eange and pasture management. 421 p. (See specifically p. 216-229.) John 
 Wiley and Sons, New York, N. Y. 
 
 89. Sampson, A. W., and H. E. Malmsten. 
 
 1935. Stock poisoning plants of California. California Agr. Exp. Sta. Bul. 593:1-90. 
 
 90. Sampson, A. W., and L. H. Weyl. 
 
 1918. Eange preservation and its relation to erosion control on western grazing lands. 
 U. S. Dept. Agr. Dept. Bul. 675:1-35. 
 
 91. Sanford, B. 
 
 1932. Some preliminary studies toward the development of a state forest fire protec- 
 tection problem. Jour. Forestry 30:616-19. 
 
 92. SCHIERBECK, O. 
 
 1931. Forestry versus game cover. Canad. Field Nat. 45 : 28-30. 
 
 93. Schmidt, W. 
 
 1929. Burned forest soil and seedlings. Der Deut. Forstwirt 11(36) :189-90; (37) : 
 193-94. 
 
 94. Sharrer, K., and K. Kurschner. 
 
 1931-32. Ein neues, raschdurchfuhrbares Verfahren zur Bestimmung der Eohfaser 
 in Futtermitteln. Die Tierernahrung (now: Biedermann's Zentralbl. fiir Agri- 
 kulturchemie und Eationellen Landwirtschaft. Abt. B.) 3:302-10. 
 
 95. Shaw, C. F. 
 
 1939. Erosion pavement. Geog. Eev. 19:638-41. 
 
 96. Show, S. B., and E. I. Kotok. 
 
 1924. The role of fire in California pine forests. U. S. Dept. Agr. Dept. Bul. 1294 : 
 1-80. 
 
 97. Sinclair, J. D., and A. W. Sampson. 
 
 1931. Establishment and succession of vegetation on different soil horizons. Hil- 
 gardia 5(7): 155-74. 
 
 98. Stallard, H. 
 
 1929. Secondary succession in the climax forest formations of northern Minnesota. 
 Ecology 10:476-547. 
 
 99. Stanley, T. L. 
 
 1932. The effect of forest fire on tourist travel. Jour. Forestry 30:608-15. 
 
 100. Steward, J. H. 
 
 1933. Ethnography of the Owens valley Paiute. Calif. Univ. Pubs. Amer. Archaeol. 
 and Ethnol. 33(3) : 233-50. 
 
 101. Stoddard, H. L. 
 
 1931. The bobwhite quail ; its habits, preservation, and increase. 559 p. Charles 
 Scribner and Sons, New York, N. Y. 
 
 102. Stoddard, H. L. 
 
 1935. Use of controlled fire in southeastern upland game management. Jour. Forestry 
 33:346-51. 
 
144 University of California — Experiment Station 
 
 103. Storie, e. e. 
 
 1937. An index for rating the agricultural value of soils. California Agr. Exp. Sta. 
 Bui. 556: 1-48. 
 
 104. Sumner, E. L., Jr. 
 
 1931. A life history of the California quail, with recommendations for conservation 
 and management. Calif. Fish and Game 21:167-256, 275-342. 
 
 105. Taylor, A. S. 
 
 1860. Sacramento Indians. Calif. Farmer 14(10) :74; (11) :82. 
 
 106. Taylor, N. H. 
 
 1938. Land deterioration in the heavier rainfall districts of New Zealand. New Zeal. 
 Jour. Sci. and Technol. 19:657-81. 
 
 107. Thompson, M. J. 
 
 1925. The effect of forest fires on land clearings and crop production. Minn. Agr. 
 Exp. Sta. Bui. 220:1-23. 
 
 108. U. S. Forest Service. 
 
 1928. Fire destroys game. Calif. Dist. News Letter 9(27) :3. U. S. Forest Service, 
 Eegion 5, San Francisco, Calif. 
 
 109. U. S. Forest Service. 
 
 1928. Wild game destroyed by fire. Calif. Dist. News Letter 9(44) :3. U. S. Forest 
 Service, Eegion 5, San Francisco, Calif. 
 
 110. U. S. Forest Service. -> 
 
 1934. Eeport on wild life within the national forests of California. 83 p. U. S. Forest 
 Service, Eegion 5, San Francisco, Calif. 
 
 111. U. S. Weather Bureau. 
 
 1928-39. Climatological data. Daily, monthly, seasonal. U. S. Weather Bureau, Cali- 
 fornia Section, San Francisco, Calif. 
 
 112. U. S. Weather Bureau. 
 
 1930. Climatic summary of the United States, Sections 15, 16, 17, 18. (California.) 
 U. S. Weather Bureau. 
 
 113. Wahlenberg, W. W., S. W. Greene, and H. E. Eeed. 
 
 1939. Effects of fire and cattle grazing on longleaf pine lands as studied at McNeill, 
 Mississippi. U. S. Dept. Agr. Tech. Bui. 683:1-52. 
 
 114. Weeks, D., A. E. Wieslander, and C. L. Hill. 
 
 1934. The utilization of El Dorado County land. California Agr. Exp. Sta. Bui. 
 572:1-115. 
 
 115. Weir, W. W., and E, E. Storie. 
 
 1936. A rating of California soils. California Agr. Exp. Sta. 599 : 1-15 7. 
 
 116. Wells, M. M. 
 
 1913. The resistance of fishes to different concentrations and combinations of oxygen 
 and carbon dioxide. Biol. Bui. 25:323-47. 
 
 117. Wright, E. 
 
 1931. The effect of high temperature on seed germination. Jour. Forestry 29 : 679-87. 
 
 18w-4,'44(8088)