May 10, 1895.] 



SCIENCE. 



511 



iu the Brazilian caiupos, or hilly region be- 

 yond the great virgin forests of the coast 

 mountains. Warmiug spent three yeai-s at 

 this place, 1863-6(). and made large collec- 

 tions of plants, which have been studied 

 and described by various specialists. Xow, 

 after nearlj- thii-ty years, the author gives 

 his general conclusions as to the flora of 

 this region, which he considers typical of a 

 great part of the interior of Brazil. The 

 mean temperature is 20.5°C, ^vith a range 

 of 3.5° to 37°C. There are two seasons— 

 di-y, from April to September, correspond- 

 ing to our winter, and wet, during the rest 

 of the year. Spring opens in August. June 

 is the coldest mouth, and December and 

 January are the warmest months, but there 

 is no winter in our meaning of the term, 

 the means of the coldest month being only 

 a few degrees Ijelow that of the warmest. 

 The annual rainfall is not known, but it is 

 considerable during part of the year, and 

 there are heavy dews in the dry season. 

 The heaviest rainfalls are in November, 

 December and January. The soil is a red 

 clay, very common in Brazil, resulting fi'om 

 the decomposition of the primarj' rocks. 

 In places cavernous limestones occur. 



There are no plains here, but only an in- 

 terminable succes.sion of hills with narrow 

 valleys through which streams have cut 

 gorges or iu which there are lakes or ponds. 

 Forests line the water courses and cover the 

 calcareous rocks. These are a meager con- 

 tinuation of the luxuriant coast forests. 

 The greater part of the country is, however, 

 destitute of trees or bears only scrubby 

 growths. These surfaces are the campos. 

 They cousist either of barren, pebbly pla- 

 teaus and Hanks of hills which are subject to 

 washing, covered with scant herbage and 

 oft.en entirely destitute of trees, or of simi- 

 lar areas bearing deeper and more fertile 

 clays and covered more or less densely with 

 herbs, shrubs and small trees. The marsh 

 and water plants form only an insignificant 



part of the vegetation, and may be left out 

 of account in this synopsis. The contrast 

 between the forest \-egetation and that of 

 the campos is very sharp, the plants of the 

 latter resembling desert vegetatiou in many 

 interesting particulars. Except iu very rich 

 parts of the campos the herbaceous vegeta- 

 tion is never dense enough to hide the hard 

 red earth. Grasses are the most im})ortant 

 part of the herbaceous covering. There are 

 about sixty species, mostly Panicnms, Pas- 

 palums and Androjiogons. All are peren- 

 nial and grow in thin scattered tufts, never 

 forming a sod. The Cyperaceie also grow 

 in the same way. The composites are rich 

 in species, especially the Yernoniete and 

 Eupatorieje. The Leguminosie come next 

 in number of species. There are 554 species 

 of herbs on the campos, but there are no 

 biennials, and the number of annuals is verj- 

 few, i. e., less than 6%. There are also 

 very few climbers or twiners although the 

 campos bears many forms intermediate be- 

 tween erect herbs and climljing and t\\'iuing 

 plants. The great dearth of annuals is 

 attriljuted to the gi"eat di-jTiess and hard- 

 ness of the soil at the time the seeds are shed, 

 to the annual fires which consume seeds 

 and seedings and may perhaps have trans- 

 formed some annuals into perennials, and 

 to the hard struggle for existence with tall 

 herbs and bushes. Herbaceous shoots de- 

 velop ordinarily in tufts and are not 

 branched or l)ut slightlj', arising in gi-eat 

 numbers from subterranean stems or roots. 

 Exclusive of certain grasses, sedges and 

 Bromeliacete, herbs with rosettes of l)a.sal 

 leaves are almost entirelj' wanting. Hori- 

 zontal rhizomes and stolons are absent and 

 horizontal ciiuline organs always remain 

 very short. Almost all of the perennial 

 Dicotyledons have a short, thick, lignified, 

 irregular, and more or less tuberous sul)ter- 

 ranean axis. Sometimes a delicate little 

 shoot only ten to fifteen centimeters high 

 arises from a tul)erous axis as large as one's 



