the origin and development of the compositÆ
TRANSCRIPT
T H E
HEW PHYTOItOGIST.VOL. XVI, No. 7. JULY, 1917.
[PUBLISHED AUGUST 9TH, 1917.]
THE ORIGIN AND DEVELOPMENT OF THECOMPOSITE.
BY JAMES SMALL, M.SC. (LOND.), PH.C.
INTRODUCTION.
THE present essay is an attempt to throw light on some of theproblems of evolution by the intensive study of a single
successful group. In 1874, when tbe section on Compositse ofBentbam and Hooker's Genera Plantarum (8) was published, thefamily included over 10,000 species; in 1897 when Hoffman com-pleted the corresponding section of Engler and Prantl's Pflanzen-familien (42) the number had risen to over 13,000; in the IndexKewensis and its supplements up to 1910 there are approximately30,000 species in the family. In such a well-defined and closelyinter-related family the genera tend to be founded on slighterdifferences than in otber families and tbe species tend tobecome elementary or completely artificial rather than Linnean, soslight are the differences which are here considered to be sufficientfor taxonomic purposes. In spite of this technicality tbe largergroups are more or less natural and the numerous intermediateforms between genera and tribes make the family a favourablesubject for evolutionary studies.
The general habit varies so mucb with climatic and geographicalsituation that the vegetative characters are of little value in thestudy of the inter-relationsbips of tbe tribes, but there are caseswhere tbese features, added to well-marked floral relationships,confirm the lines of development already indicated. The physiology(irritability of the pollen-presentation mechanism, latex, etc.) andthe cytology can be shown to develop in conjunction with thechanges in floral structure, but ultimately the morphology of theflower and capitulum is the real test of relationship and development.The study of the details of tbe flower in tbis homogeneous group
158 James SmaU.
becomes, therefore, particularly important, and as there is leastvariation in the essential parts of the flower these must be consideredmore important than the very variable non-essential parts. Thiswas recognised by the first synantherologist, Cassini, and isconfirmed by Bentham, the only other botanist who has studied theCompositse sufficiently profoundly to express a sound and independ-ent opinion.
After dealing with the history of the Compositse we shall,therefore, consider the variation in the form and structure of thestyles and stamens. As these are closely connected in the work ofpollen-presentation the development of that mechanism, including itssusceptibility to stimuli, is of first importance.
The variations in the corolla are considered in relation to thedoral development and insect visitors ; the variations in the pappusare considered in relation to the component parts of the trichomeand the dispersal of the seeds. The characters of the receptacleand involucre are considered in relation to their development asindicated by closely allied forms of receptacle and involucre and inrelation to the lines of evolution suggested by the examination ofless variable parts of the capitulum. The various phylogenetic linesindicated by the critical examination of these floral characters aretested by the geographical distribution of the groups and thephyllotaxis and cytology of the family are found to follow the phylo-genetic lines established by the preceding investigation.
In the discussion of the inter-relationships of the tribes andgenera, use is made of the modern theories of heredity, evolutionand geographical distribution. Some of these may be regarded asrather speculative, but their use is considered justified by the coherentaccount of the evolution of the family which results. Lines forfurther research are abundantly indicated and correction or confirm-ation of the views expressed will be sought along the various lines ofphysiological relationships, cytological and geological developmentand the details of geographical distribution.
Origin and Development of the Compositce. 159
CHAPTER I.HISTORY OF THE CLASSIFICATION OF THE COAlPOSITyE.
In the history of synantherology the classifications of thetaxonomists have been in the majority of cases moreor less artificialattempts at grouping related genera and only a few have made anyattempt at expressing the affinities of the tribes and sub-tribes. Ontheotherhanda number of studentsoftheCompositse have expressedviews concerning the relationships of the tribes, hut these viewshave heen founded in some cases on somewhat limited enquiries intothe anatomy and morphology of a number of genera.
It has heen found convenient, therefore, to discuss the historyof the grouping of the genera into tribes and the history of thegrouping of the tribes according to their supposed relationships inseparate sub-sections.
A. TAXONOMY.
The present sub-section deals not so much with the precisedetails of the classification of the family as with the evolution ofthe idea of the family as such and of the ideas of the differenttrihes as groups. It is an attempt to trace this evolution from thefirst more or less sub-conscious grouping of Soitchus, Cichoritim andother Cichorieae hy Theophrastus to the modern conceptions ofthe divisions and sub-divisions of this immense and homogeneousmass of species.
Aitcietit Systems.Theophrastus. Circa B.C. 320.
The origin of the idea of two of the tribes of the family, theCichorieae and Cynareas, seems to be lost in antiquity like the originof some other very natural groups, which, as Greene (37) says, musthave heen recognised hy primitive man from the earliest times.They are distinguished very clearly by Theophrastus in his Enqttiryinto Plants (83). The chicory-like plants (x'X''V"Ji'S'?s) discussedin Book VII, Chapter XI, include Cichoritivi, Hypocliceris, Taraxacumand Picris, all descrihed and distinguished from one another andgiven, moreover, only as examples of the class. The distinguishingcharacters of the grouparegiven clearly andconcisely. Theophrastusalso mentions in Book 1, Chapter X, the pine-thistle and " all theplants which belong to that class," the d/cavwSiov or u.Ku.vOwhmv.As he mentions the Acanaceae again more than once it is obviousthat the recognition of the Cynareae as a group was to Theophrastusa commonplace. In Book I, Chapter XIII, he also describes agroup of plants distinguished by having the flower on the top of theactual seeds and with one flower attached to each seed; as these
i6o James Small.
are described as " all thistle-like plants " it seems tolerably certainthat in addition to Cichoriese and Cynare^ this perspicaciousGreek recognised the Compositse as a class. The only other largeobviously natural family which corresponds to this definition is theUmbellifer^ and this was separated as the Perulaceje.Dioscorides. Circa A.D. 65.
According to Smith's Classical Dictionary, Pedacius Dioscoridesprobably lived in the second century A.D., but Burgess (16) givesevidence for a date about 65 A.D. for the writings of this author ;Adanson (1) gives 50 A.D. as the date. The variations in thedifferent editions of his Materia Medica are notorious but as earlyas 1555 the grouping of Sonchus, Cichoriitm, and Chondrilla byDioscorides was amplified in the annotations of a Castillean editionby Dr. Andr6s de Laguna (23) who noted the similarity to the above-mentioned genera of Scariola, Picris and Taraxacttm.
In the edition of 1598 by Bauhin of Matthioli's Dioscorides (24)other groupings of Composites are to be noticed, thus Lib. Ill,Cap. XIX, deals with Lencantha, identified as Carditus Marice byBauhin and the subject of Cap. XXI is identified with Carduusstellatus. Chapter 109 of the same book deals with Tussila^oFarfara, 110 and 111 with forms oi Artemisia and 112 with A>nbrosia;similarly Antheinis, Parthenittm {=Matricaria) and Buphthali)iitm{Bellis, according to Bauhin) are dealt with in chapters 137-139. Itis possible that this grouping of Antheinis and Matricaria forms theflrst indication of the Anthemideae and this is rendered more probableby the development of the group by Brunfels.
16th Century.Following upon Dioscorides comes a gap in the history of
botany in general of about fifteen centuries. During the 16thcentury various commentators on Dioscorides, such as Brunfels (15),Valerius Cordus (20) and Puchs (31) established and extended theAnthemidese while retaining the Cichorieae and Cynare^. Lobel (60)extended these three groups and classed a few genera of the Asterezetogether. Caesalpino (17) was the flrst to group the Compositeas a whole. His " Herbacese pluribus seminibus" are divided intothe " Ant'hemideffi " and " Cichorace^ aut Acanace^." The" Anthemide^," however, include all the Composites not in theCynarese or Cichoriete, while the Anthemidea: as recognised by theearlier writers is considerably obscured.
17th Century.Bauhin (6) fails to recognise the Compositse as a family but
retained the groups previously recognised, i.e., the Cichorieae
Origin and Development of the Compositcr. i 6 i
Cynarese, Anthemideae and Astere^, and in addition he classedseveral of the genera of the Inuleas together. A decade later Gerard(33) advanced still further with a suggestion of the Calendule^ as agroup and the consolidation of the Inuleas in addition to the groupsof Lobel.
In the latter half of the century Morison (65) and Ray (72-73)again recognised the Compositae as a family and agree in retainingthe Cichorieae and Cynarece while confusing all the other tribes. Theremaining authorities of the 17th century, such as Christopher Knaut(48), Rivini (75), Hermann (39) and Tournefort (84) agree in regardingthe Composites as a family with the Cichorieae as a distinct section,but confuse the other tribes. Knaut seems to be the first to use thename Compositae, Ray's designation is " Composito flore." Hermannclasses the group as " Gymnomonospermas Compositae."
18th Century.The dominating systematist of this century, was, of course, Linn^,
but in his three attempts (59) at the classification of the Compositzehe made no progress whatever. Indeed, his first arrangementincluded so many external genera, such as Globularia, Protea andLencodendron that he can scarcely be said to have recognised thefamily at all. His " Syngenesia " in the Sexual System includeLobelia, Viola and Impatiens ! The Ambrosia group are separatedfor the first time in his Natural System.
Linne was preceded by a group of systematists whose ideas onthe Compositae varied very much. Boerhaave (12) gives thecharacter " staminibus propriis coalitis in tubum" for the firsttime and has one section " semine aculeis donato " which fore-shadows the Helianthe^. Christian Knaut (49), Ruppius (76),Magnol (62), Ludwig (61), Allionius (2) and Necker (66) madeattempts somewhat similar to the systems of Tournefort and Rivini ;in some cases the presence or absence of rays was used as a diagnosticcharacter.
The first botanist to consider the Compositae alone was Vaillantus(85) whose system bears a distinct resemblance to that of Boerhaave,with an amplification of the Helianthes. Pontedera (71) gives asystem very similar to that of his contemporary, Vaillantus, but withsuggestions in his grouping of the genera of several tribes such asthe Senecioneas, Calenduleae, Astere^e and Helenieas, which werenot recognised by that systematist. Berkhey (9) and Meese (64) alsopublished dissertations on the Compositas but these show little or nooriginality.
t62 James Small.
Adanson (1) reviewed previous systems and places the Campan-ulaceae next the Composite. He recognised some of the previouslydistinguished trihes (vide Table I) and followed Linne in separatingthe Anthrosia group but retained it within the family. Jussieu (46)makes the " anthers connatse " a diagnostic character of the familyand foreshadows the Arctotide^ and Mutisieas in his Cinarocephalasanomalae. In his arrangement of the genera Gaertner (32) followsPontedera in the characters (receptacle, pappus and ray) used andarrives at a similar result, recognising the Inuleae and Anthemideaein addition to Pontedera's groups.
Early 19th Centitry.
In the beginning of this century Richard (74), Batsch (5) andLagasca (51) followed their predecessors with little or no originalityin the Compositse, except that Rich used the divided or undividedstyle to characterise his two sections of the Synantheria.
Cassini, 1813-7834.
The founder of synantherology and the master in whose footstepsall subsequent students of the Composite have followed wasHenri Cassini. As the result of a systematic analysis of all thefloral characters he arranged the genera of the family into a seriesof 20 tribes with sections and sub-sections (vide 18, Vol. Ill, Synopsis)which have been retained as units during the various minor changesmade by later systematists. There is no detail of morphology whichescaped the notice of Cassini, and the present intensive study of thefamily by including physiology and geographical distrihution canmodify only slightly the grouping of the trihes. Even thenCassini had already indicated the existing relationships which will heemphasised later.
Cassiniaii Period, 1820-1917.
There have been three chief modifications of Cassini's system.Lessing (55) hy using only the style characters reduced the numberof tribes to eight but retained practically all Cassini's sub-tribes. DeCandolle (21) followed Lessing with the addition of many originaldivisions of the sub-tribes. Bentham (8) returned to Cassini's systembecause he appreciated the value of the characters of the stamensand he acknowledges this return (7) hut claims to have arranged thefamily before he was aware his system was so similar to that of Cassini.The present system is, therefore, one reached independently by theonly two botanists who ever studied the family sufficiently thorough-ly to be able to speak with authority.
Origin and Development of the Compositce. 163
Other systematists follow these four more or less closely :Bartling's system (4) is frankly Cassinian ; Link (58) followedCassini with several retrogessive variations; Lindley neglectsCassiniand Lessing, returning to Jussieu (56) and later (57) following DeCandolle. Endlicber (29), A. de Jussieu (47) and Eichler (27-28a)all follow De Candolle. Payer's arrangement (70) is a curiousatavistic return to the early 18th century and Baillon (3) fusesseveral pairs of Benthamian tribes. Hoffmann's arrangement (42)shows no originality. One or two obvious sub-tribes are sunk andminor suggestions made by Bentham are carried out. Hoffmanngives an artificial sub-division of the Cichorieee, but Engler andGilg (30) returned to the Benthamian arrangement. Wettstein (89)
follows Hoffmann.Floras.
The Floras of any given period usually follow the current system.Thus Meese (63) and Nuttall (69) follow Linne; Kunth (50) followsCassini with slight variations; Gray follows first De Candolle (34-35)and then Bentham (36). Boissier (13) follows Bentham.
American Systems.The Americans show a tendency to raise the status of thegroups:
thus Jepson (45) raises the Ambrosinje and Madinse to the rank oftribes ; Britton and Browne (14) raise the Ambrosinte and Cichorieseto the rank of families ; J. K. Small (79) follows Britton and Browne,and Bessey (11) raises all the tribes and the Ambrosinae to the rank offamilies, making fourteen in all.
Special Memoirs.Don (25) and Schultz-Bipontinus (78) attempted classifications
of the Cichoriese. Nees (67) and Burgess (16) have dealt with theAsters and Delpino (22) proposed a new sub-division ol the Senecion-ideffi of Lessing, but tbese isolated memoirs have had no effect onthe general classification of the family.
Conclusion.From Table I it will be obvious that the history of the classifi-
cation of the Compositas is clearly divided into two epochs, pre-Cassinian and Cassinian, witb little or no advance from Theophrastusuntil the few decades immediately preceding Cassini, in which someslight indications are given of the origin of the ideas of a few ofthe tribes afterwards defined by the master. Tbe half centuryfollowing Cassini was marked chiefly by the burying of his classicmemoirs by Lessing and De Candolle and tbe present Benthamianperiod is nothing more than a return to the teaching of the greatestof all synantherologists.
164 James Small.
TABLE I.
Tribes recognised frojn Theophrastus to 1915.
Date.
B.C.3MA.D.65
153215401542157015831623163316801682-8616871690169017001710-20171617181718-211720172017371736-38
,,
175717601760-61176317891791179118U11802181618181813-3418201829183018321830361836-381836-401860187318731875-80188218311857188418971898190119011903-131915
Author.
Theophrastus
DioscoridesBrunfelsCordusFuchsLobelCaesalpinoBauhinGerardMorisonR a yKnaut (1)RiviniHermannTournefortBoerhaaveKnaut (2)KuppiusVaillantusPontederaMagnolLudwigLinn6
,,
AllioniusBerkheyMeeseAdansonJussieuGaertnerNeckerRichBatschLagascaNuttallCassiniKunthLinkBartlingLessingLindleyDe CandolleEndlicherPayerBenthamBoissierEichlerBaillonGray
. 1
HoffmanBritt. & BrownWettsteinJepsonSmall, J. K.Bessey
K
oD.
O
+
+
_j_1
+
+
-
+
'(+
+_l_
( + •^)( + •)<•)
(+*)-f-f
-1-- j -
+
( + J( + "M*!( + •^1( + •K''
(+*]+ •»*•
( + * '
( + * ;
+ -*f
(+•>*•)
pi
Eou(U
_l_• >
+-_(_
+
oc_
+
+
++
+
i
+
ic
"3
-|-
•3
-^
^
+++
Kj=
c
X
+
?
+
B
X
•p
_f_
_j_
,
7]
Ant
liem
ide^
e. 1
-[-
+
'
]
Hc0
Ofc01
rJH
*
"T
+
+
+
_[__l_
+
(U
c
_
"''"*""*"
"'"
+
i
-
+
+
+
'
'^
"'"
+
"^4-+
+
+
cU
+
- j -
_ L
' " "
++
+
+• * •
" * "
•1"
++
+++
+
+
+
H<u
0
"o0
+
_l_
^
4^
• >
" * "
' \ ^
• )
+
"I"++
+
+
+
+++++
+
+
+- 1
-f Family or tribe recognised as a group.-I-* Ambrosiaceje as a distinct family.( + *) Ambfosiacea a8 a tribe or sub-tribe.
Origin and Development of the Compositce. 165
B. PHYLOGBNY.
Beyond vague ideas of a common aneestry for some of thegenera obviously very much alike, the pre-Cassinian writers on theCompositae had apparently no thoughts of the evolution of onegroup from another within the family. It is, of course, well knownthat many of them were vehemently opposed to all evolutionarydoctrines. As this question of creation or evolution is a general oneand as Cassini was the first to distinguish the tribes sufficientlyclearly to form any correct idea of their relationships the presentaccount will be given in chronological order from Batsch and Cassinionwards. Another reason for the chronological order is that itemphasises the penetration of Cassini and the subsequent neglectof his valuable work by succeeding writers to the great detrimentof synantherology.Batsch, 1802.
In his Tabula affinitatum (5) Batsch discusses the affinities ofeach of his three groups of Composites. He considers the Lepid-ocephalae (Cichoriese) to be connected with the Cinarocephalte(Cynareae and Mutisiese) through Scolymns carduifortmis andBarnadesia, with the Campanulace^ by the aggregation of the flowersand by the latex and with the Cucurbitacese by the bitter juice anddisagreeable odour. The Cinarocephalte he considers to be connectedwith the Valerianaceae and Proteaceae by theexalbuminous seed andaggregation of the flowers; similarly the Corymbiferae are said to haveaffinities with the Campanulaceae.Cassini, 1826.
The form chosen by Cassini for the expression of his views onthe affinities of the tribes of the Compositse is one seldom used butvery useful. In Plate I of the Opuscitles (18) he places the nineteentribes each in a circle and the circles in an ellipse. Fig. 1 is fromCassini's diagram in the Opuscules, which he describes as a "tableauexprimant les affinit^s des tribus naturelles de la famille desSynanth^rdes." The Boopidees are the Calyceracese of modernsystems. The interesting points in relation to the present study ofthe Compositse are the affinities indicated between the Senecioneaeand the Eupatoriese, the Senecionese and Asterese, the Senecioneaeand Anthemidese, the Senecioneae and Mutisie£e, and between theInulese and Cynareae.Lessing, 1832.
On page 435 of the Synopsis (55) Lessing gives a table of thetribes and sub-tribes of which he remarks " Analogiam subtribumsingularum videre licet e tabula sequente." An examination of thisarrangement (Table 11) shows that Lessing followed Cassini in his
166 James Small.
FIG. 1. Cassini's scheme of relationships of the tribes of Compositse.
recognition of the affinities of the Senecioneae with the Anthemideae,Astereae and Mutisiese and that he correctly appreciated the similaraffinities between the Senecioneas, Calenduleae and Arctotideae. His
Origin and Development of the Compositus. 167
VII
Iri
bes
of
Co
mp
osi
tce.
VI
VII
on
ship
s of
the
VT
AB
LE
II.
Les
sin
g's
sch
eme
of r
elat
II
III
IV
-
0>(J
3(I)3ca
Z
Sen
ecio
nide
se
KQJ
'oL.
05
Eup
ator
iace
ae
55cu
(U
CQ
OX
O
0)
CQ
"m
3
s
73'5
c
6
Sen
ecio
nese
Tag
etin
ese
Hel
enie
seC
brys
antb
emea
E
H
| |
Mut
isie
ae
M
.2 -'-' A uS S Si's
Ley
sser
ieas
3C
IU
K0)•gCQ
X<J<JOS
CQ
Kc
'So
CQ
en
3
M01cco.X
o
CQXo,c
a0>
X
cCQ
co
Xu5
M
%-za a[I. J
Rel
bani
eae
1CQ
XXa .3
xE
.9"
H
enO
XCJ
K0)
1xe
Mel
ampo
dies
e
Nas
sauv
iese
K73o
7 3 CQC X_« D.
K.-2^en -wa . 05
IIWO
01
oa.3U
'Soc1.fj
H
" '3 o^S u
3 « N o mt3 o o P-°
Car
duin
ea;
U
au
« £
168 James Small.
placing of the Othonnese in the Cynaroideae and theinese in the Eupatoriese marks the affinity of the BenthamianSenecioneae with these two tribes, as the Othonneae and Tussilagineaenow form sub-tribes of the Senecionese. His grouping of the Inule^and Leysserese, Tarchonantheee and Gnaphalie^, BuphthalmejeandRelhanieae is interesting as all these are sub-tribes of the Inuloideaof Bentham. Various other groupinjJs are of minor interest but willbe obvious to the synantherologist.
Delpino, 1871.The stress laid upon the anemophily in Artemisia by Delpino
(22) has been shown by Bentham (7) to have led to erroneousconclusions. His views on the origin of the family as shown in TableIII are interesting on account of the recognition of the affinitybetween the Lobeliaceae and Compositae. It is necessary to rememberhowever that his Senecionideze is that of Lessing and includes theHeliantheEe, Helenieae and Senecioneae.
TABLE III. Delpino's scheme of the derivation o/Absinthium.
AbsinthiumA
SenecionideaeA
CompositECA
LobeliaceEeA
Campanulacese.
Bentham, 1873.Bentham followed Cassini very closely in his expression of
the affinities of his tribes, as will be seen from Fig. 2, which is thearrangement given in Plate II. of the Notes (7). This diagram failsto indicate the close affinity of the Senecioneag and the EupatoriaceiEthrough the Tussilagineae, recognised by Cassini and certainly notobliterated by the removal of that subtribe from the latter to theformer tribe by Bentham. Similarly it omits the connection betweenthe Cynareae and Inuleae noted by Cassini It marks clearly, how-ever, the affinities of the Senecioneas with six other tribes.
In addition to the above diagram Bentham gave expressionto other views on the history of the family in the text of the Notes
Origin and Development of tlie ComposiUc. 169
(pp. 481 sqq.). He considers the Helianthoidese to be the mostancient tribe on account of the foliaceous involucral bracts, aristatepappus and free anthers of some genera. Prom the ancestralHelianthoidese by the regular development of a tubular corolla andthe substitution of a pappus for a reduced calyx he supposes theEupatoriacese arose in America, the Vernoniaceas in the New andand Old Worlds, the Cynaroidese in the northern and the Mutisiacesein the southern hemisphere. From the same plexus, by a reductionof the corolla, a disappearance of the stamens and an obliquedevelopment of the corolla in the outer florets, he supposes themodern Helianthoideas arose in both West and East, and also theHelenioidese in the former hemisphere and the Anthemideze in the
FIG. 2. Bentham's scheme of relationship of tbe tribes of CompositaE.
I 70 James Small.
latter. By reduction in the involucre and development of a setosepappus the Asteroideae, Senecionidese and lnuloidese are said tohave arisen in both Old and New Worlds from this Helianthoidplexus. Finally by a unilateral development of all the corollas, areduction in the involucre and the development of a pappus hesuggests the Cichoriese arose also from the ancestral Helianthoideze.It will be noticed that these opinions scarcely coincide with thoseexpressed in the diagram (Pig. 2), and they will be discussed laterin the light of the present investigation.Vuillemin, 1884.
The most comprehensive study of the anatomy of the Com-positas is Vuillemin's Tige des Composees (86), an investigation" de la valeur des caractdres anatomiques au point de vue laclassification des veg^taux." He concluded that although anatom-ical characters may be of value in the identification of fragmentsof plants such as drugs and fossil debris or in indicating the affinitiesof species within a genus or in special cases even genera withintribes, they were of no value as a basis for classification in thelai'ger groups of the Compositse.Hildebrandt, 1887.
VuiUemin's conclusions were controverted three years later byHildebrandt (40), who, however, confined his investigations to theAmbrosiaceffi and Senecionidese (of Lessing). He reclassified thegenera he examined on the anatomy of the stem, but his sub-sections as he characterises them show many variations in theiranatomy, and there is an example of almost each type of structurein most of his subtribes.Nichols, 1893.
After studying the achenial hairs of the Compositse Nichols(68) decided that they were of no use as tribal characters." Within the genus, however, the characters seem to be moreconstant, and might, perhaps, be made of value in the determinationof species."Hock, 1898.
Dealing with orders and not with families Hock (41) derivesthe Campanulat^ from a line showing epigyny ; the main line givesa Rubialian line near the base from which a line leading to theUmbelliflorse is given off, the main line culminating in theCampanulatse.Col, 1899.
Prom his study of the laticiferous tissue and other secretory
Origin and Development of the Compositce. 171
canals in the Compositae Col (19) considered that the Cynare^ hadarisen from the " Radices" and that the Calenduleje, Arctotideaeand Cichorieae were developed on a line of evolution parallel to thatwhich gave the thistles. His views are expressed thus :—
Cynarees
-Calendulees Arctotidees Liguliflores
Hallier, 1905.The Campanulatje according to Hallier (38) came from the
Gymnosperms via the Polycarpicse, Magnoliacete, Aristolochiacejeand Passiflorales. He discusses briefly tbe Campanulaceas,Calyceracese and Compositse, considering the last to be thehighest expression of several of the tendencies shown by the Campanulat£e.
Lavialle, 1912.After an elaborate and extensive investigation of the develop-
ment and structure of the achene Lavialle (52) gave expression toan opinion upon the affinities of the Cichoriese, Cynare^ andMutisieae. The examination included 298 species, 65 genera andmost of the subtribes in the above-mentioned groups. Lavialledid not, however, extend his observations to the rest of the Com-positse so far as I bave been able to ascertain at present. Hisviews are given in tbe form of a diagram, Fig. 3, but tbeir value is
FIO. 3. Lavialle's scheme of the affinities of the Mulisicic. C)iiarca;,and Cichoriese.
172 James Small.
somewhat reduced by the limitation in the number of tribesexamined.Wernham, 1913,
In his account of floral evolution in the Sympetalse Wern-ham (88) does not consider groups below the rank of familybut he supports the view that the "Compositse have arisen from theCampanulacete, an affinity which was recognised early in the historyof the two families. The most important of Wernham's contribu-tions is contained in the section on the determination of plantaffinities in his " Summary and Conclusion " to which reference willbe made later.Dufour, 1907, and Lebard, 1913.
These two investigators (26 and 53) laid considerable stressupon the form of the cotyledons in the Cichoriese, and Dufourproposed that the tribe should be divided on this character intotwo subtribes, the Brachycotyldes (type Lactuca with short, broadcotyledons) and the Leptocotylees (type Scorzonera with long, narrowcotyledons). Dufour's work was carried further by Lebard, whodistinguishes a form of cotyledon intermediate between the shortand the long forms and gives a phylogenetic diagram shown inTable IV.
TABLE IV.
Lebard's phylogenetic scheme of tlie Cichoriece.
TUBULIFL0RE5 TUBULIFL0RE5
$<:oKionm ms iPOOOSPERMUM
CiCHORlDEESJAUMACUM
CiCHORiuM
ILACTUCA
Origin and Development of the Camposilcc. i 72,
He regards the Leptocotylees as primitive and the Brachycotyl^esas derived along two lines of evolution from the Tubuliflorse.
Lee, 1914.After an investigation of the seedling anatomy of about 50
species of Compositae belonging to most of the tribes Lee (54)concluded that the classification in this family could receive noassistance from seedling anatomy. As he found all the types ofanatomy which occur in the family in closely related species, andeven in different specimens of the same species it is obvious thathe was right in his conclusion that seedling anatomy is of no valuein dealing with questions of affinity within the Compositae. This isone more example of the failure of anatomy to give phylogeneticcharacters in the Compositse, and agrees with the facts given byVuillemin and Hildebrandt, although not with the latter's con-clusions.
Bessey, 1897, 1915.ln his presidential address to the Botanical Society of
America in 1897 (10) Bessey placed the Asterales at the summit ofthe Rubialian line of evolution. The four families included in theAsterales were supposed to be derived from one another thus :—•Valerianacese—>• Dipsacese—> Calyceracese—> Composite. Thesame author in 1915 (11) enunciated various dicta for determining
TABLE V.
Bessey's phyletic lines for the tribes of the Coiitpositce.
- /\RCTOTiDACE/L
" 294 '" ^95->' ^96-Am HE mi DACE/I" $97'5tnEaoniD
" 208-CARDUACt/E
I 74 James Small.
the phyloj»eny of flowering plants. These dicta, however, applyonly to certain groups in each case. He raises the fourteen tribesof the Compositas or Asterales to the rank of families, and rej^ardsthe Helianthaceae as the primitive group from which two principalphyletic lines have arisen, culminating in the Eupatoriacese on theone hand and the Lactucace^e on the other. His views may beexpressed as in Table V.
Hutchinson, 1916.The Helianthese are also regarded as the most primitive tribe
by Hutchinson (43) who considers the aristate or paleaceouspappus and the receptacular paleas as primitive characters, andwho also considers (44) the structure of tbe pappus in the Com-positte as " perhaps one of the most important features in theconsideration of the phylogeny and affinities of the genera of thisinteresting family." This is very different from Bentham's view ofboth pappus and receptacular paleje, the inconstancy of which ingenera and sections led to that synantherologist to consider thesecharacters as of use only when supplemented by other and moreocnstant features.
Small, 1915-16.The present writer has made various suggestions as to the
affinities of the Compositse and its constituent tribes. The first (80)was made after a study of the variation in styles and stamens andthe elucidation of the function of the appendages of these organs. Thechief point made was the establishment of similar lines of evolutionfrom the consideration of the styles and of the stamens. As the workalong these lines has been revised and extended it will be enoughat this juncture to note that the Senecioneae were considered theprimitive group. In a subsequent study of the corolla (81) it wassuggested that the Cichorieae arose directly from the Senecioneaeby a comparatively small mutation. Other observations on thedisplacement and multiplication of the ovules in Senecio vulgaris (82)controverted the removal by Warming (87) of the Calyceraceaefrom the neighbourhood of the Compositce. These studies havebeen continued, and the present account is the result of theextended investigations.
BIBLIOGRAPHY I.
1. Adanson, M. Families des Plantes. Paris, 17G3.2. Allionius, C. Stirpium Prjecipuarum Littoris et Agri Nicseensis. Paris
1767.
Origin and Development of the Compositce. \ 75
3. Baillon, H. Histoire des P l an t e s . Tom. VIII , Par i s , 18S2.4. Bar t l ing, F . T. Ordines Na tura les P lan ta rum. Gott ingen, 1830.5. Batscb , A. J . G. C. Tabula affinitatum Regni Vcgetabilis. Vinaria:, 1802.6. Baubin, C. Pinax, tbeatr i botanici . Basi lex . 1G23.7. Bentbam, G. Notes on tbe Classification, History and Geographical
Distribution of Composita2. Jour . Linn. Soc. (Bot.) ,Vol. X n i , 1873.
8. Ben tbam, G., and Hoolter, J . D. Genera P l an t a rum. Vol. 11., P t . I.London, 1874.
9. Berkhey, J . le F . van. Exposito Charac t e r i s t i c s Structural Florum quidiacuntur Compositi . Lugden, 1760.
10. Bessey, E. A. Pbylogeny and Taxonomy of Angiosperms. Bot.Gazet te , Vol. XXIV, 1897.
11. ,, ,, The Pbylogenet ic Taxonomy of Flowering P lan t s .Annal. Missour. Bot. Gdn., Vol. II, Nos. 1 and 2,1915.
12. Boerhaave, H. Index P l a n t a r u m liorti Lugduno-Batav i . Lugden, 1710.
13. Boissier, E . Flora Oriental is . Vol. I l l , Geneva, 1875.
14. Bri t ton, N. L., and Brown, A. An Il lustrated Flora of the NorthernUnited Sta tes , .S.C. New York, 1898.
15. Brunfels, O. Herbarum viv^e eicones. 1532.
16. Burgess, E. S. I—History of Pre-Clusian Botany in its Relation toAster. 11 —Species and Variations of Biotian Asters .Memoir Torrey Bot. Club, Vol. X and XIII, 1902,1906.
17. Caesalpino, A. De Plant i s . Lib. XVI, Florence, 1583.
18. Cassini, H. Opuscules Pbytologiques. Tom. M I , 1826, Tom. I l l ,Par is , 1834.
19. Col, M. Quelques recbercbes sur l 'apparcil secreteur des Composees.Jour , de Botanique, Vol. I l l , 1899.
20. Cordus,Valerius. Historia P lan ta rum, M. S. S., 1540. Opera Omnia,15Gl.
21. De Candolle, A. P . P rod romus Sys temat is Natural is . Vol. V VII, Par i s ,1836 38,
22 Delpino, F . Studi sopra un lignaggio anemofilo delle Composte. Firenze,1871.
23. Dioscorides, Pedacius . Materia Medica. Traduz . Griega-Castel lana.IUus. by Dr. A. de Laguna. Anvers, 1555.
24, ,, ,, Materia Medica. P . A. Matthioli 's Commentari i ,by C. Bauhin. Basileai, 1598.
25 Don D An At tempt at a New Classification of tbe Ciehoriaceas. Edin .New Pbil . Jour . , Vol. VI, p. 305, 1828.
26 Dufour L. Observat ions sur les affinitis et devolution des Cbicoracees.Compt. Rend. Tom. , CXLV, 1907.
27. Eicbler , A. W. Blu tbendiagramme. Leipzig, 1875.28 Syllabus der Vorlesungen uber Phanerogamen Kunde.
Keil, 1876.OOQ Syllabus der Vorlesungen iiber specielle undmed. pbarm.
Botanik. Berlin, 1880.
29. Endlieber, S. Genera P l an t a rum. Vindobona;, 1836-40.
30 Engler A., and Gilg, E . Syllabus der Pflanzenfamilien. Berlin, 1912.
31. F u c b s , L. Historia St irpium. Basi led, 1542.
32 Gaer tner , J . De Fruct ibus et Seminibus P lan ta rum. Tubingen, 1791.
33. Gerard , J . Historie of P lan t s , Bk. II. London, 1633.
34 Gray Asa, and Torrey, J . Flora of North America. New York, 1838-40.
' fi Gray Asa Manual of Botany of Nor thern United Sta tes . New York^ ' 1857.
176 Origin and Development of the Compositor.
36 ,, ,, Synoptical Flora of North America. New York, 18H4.37. Greene, E. L. Landmarks of Botanical History. Pt. I. Prior to 1562A.D.
Smithsonian Miscell. Collect, Vol. 54, 1910.38. Hallicr, H. Verwandschaftverh. der Tubifloren und Ebenalen den poly-
phj'let. Ursprung der Sympetalen u. Apetalen.Naturwiss. Verein-Abhand. Bd. XVI, Hamburg,1903.
39. Hermann, P. Florae Lugduno Batavse. Lugden, 1690.40. Hildcbrandt, F. Bcitrage zur vergleich. Anatoniie der Ambrosiaceeii
und Senecionideen. Marburg, 1887.41. Hock, F. Kui'ze Bemerkungen zur Systematik der Kormophyten.
Botan. Centralbl. (Uhlworm) Bd. LXXVI., p. 171,1898.
42. Hoffmann, O. In Engler and Prantl's Die natiirlichen Pflanzenfamilien.Teil IV, Abt. 5. Leipzig, 1897.
43. Hutchinson, J. Aquatic Compositje. Gardener's Chronicle, Vol. LIX,p. 305, 1916.
44. ,, ,, Notes on African Compositje, II. Kew Bulletin XXXI,1916.
45. Jepson, W. L. A Flora of Western Middle California. California, 1901.46. Jussieu, A. L. de. Genera Plantarum. Paris, 1789.47. Jussieu, A. de. Taxonomie. Paris, 1848.48. Knaut, Christopher. Enumerat. pi. cire. Halam. Leipzig, 1687.49. Knaut, Christian. iMethodus Plantarum genuina. Leipzig, 1716.50. Kunth, K. S., Humboldt, A. V. and Bonpland, A. Nova Genera et Species
Plantarum. Tom. 4, Paris, 1820.51. Lagasca, \V. Genera et Speeies Plantarum. Madrid, 181ti.52. Lavialle, P. Recherches sur le developpement de l'ovaire en fruit chez
les Composees. Ann. d. Sci. nat., Bot., 9e ser., t.XIV, 1912.
53. Lebard, P. Remai'ques sur les affinites des principaux genres du groupedes Liguliflores. Compt. Rend. T. CLVII, 1913.
54. Lee, B. Seedling Anatomy of Sympetalas, II, Compositse. Annals ofBotany, Vol. XXVIII, 1914.
55. Lessing, C. F. Synopsis Generum Compositarum. Berolini, 1832.56. Lindley, J. An Introduction to the Natural System of Botany. London,
1830.57. ,, ,, A Naten-al System of Botany. 2nd edit., London, 1836.58. Link, H.F. Handbuch zur Erkennung der nutzbarsten undam haufigsten
vorkommenden Gewachse. Berlin, 1829.59. Linn6, C. Classes Plantarum. Lugden, 1738.60. Lobel, M.de. Stirpium adversaria Nova. London, 1570.61. Ludwig, M. C. G. Deflnitiones Plantarum. Leipzig, 1737.62. Magnol, P. Character plantarum novus. Monspelii, 1720.63. Meese, D. Flora Frisica. Franeker, 1760.64. ,, ,, Syngenesia. Leeuwarden, 1761.65. Morison, R. Historic Plantarum. Oxford, 1680.66. Neeker, N. J. de. Elementa Botanica. Neowedse ad Rhenum, 1791.67. Nees, C. G. Genera et Species Asterearum. Nuremberg, 1833.68. Nichols, M.A. Achenial Hairs of Composite. Bot. Gaz., Vol. XVIII, 1893.69. Nuttall, T. The Genera of North American Plants. Philadelphia, 1818.70. Payer, J .B . Lemons sur les Families Naturelles des Plantes. Paris, 1860.71. Pontedera, J. Methodus Compositarum. Dissertationes XI, Botanies
Patavini, 1720.
Rhizophidium acuforme (Zopf) Fisch. 177
72. Ray, J. Metbodus Plantarum. London, 1682.73. ,, ,, Historia Plantarum. London, 1686.74. Richard, L. C. in "Catalogue des Plantes du Jardin medicinal de Paris,"
by F. Marthe, 1801.75. Rivini, A. Q. Ordines Plantarum. Leipzig, 1690.7(). Ruppius, H. B. Flora jenensis. Frankfort, 1718.77. Schultz Bipontinus, C. H. Beitrage zur Kenntniss der Cassiniaceen.
Flora, Bd. 35, p'. 129, 1852.78. ,, ,, ,, Beitrage zum Syskema der Cichoriacean.
Di.irkheim, lS6fi.79. Small, J. K. Flora of South Eastern United States. New York, 1913.80. Small, J. The Pollen-presentation Mechanism in the Compositre.
Annals Bot., Vol. XXIX, 1915.81. ,, ,, Notes on tbe Corolla in tbe Compositre. New Phytologist,
Vol. XV, 1916.82. ,, ,, Anomalies in tbe Ovary of Senecio Vulgaris, L. Annals Bot.
Vol. XXX, 1916.83. Tbeophrastus. Enquiry into Plants. With Eng. trans, by Sir A. Hort.
London, 1916.84. Tournefort, J. P. Institutiones Rei Herbarise. Paris, 1700.85. Vaillantus, S. Methodus Compositarum. Memoires de l'Acad. Roy.
des Sei., Paris, 1718-1721.86. Vuillemin, P. Tige des Composees. Paris, 1884.87. Warming, E. Observations sur la valeur systematique de I'ovule.
Mindeskrift f. Japet. Steenst., 1913.88. Wernham, H. F. Floral Evolution : with particular reference to the
•Sympetalous Dicotyledons. New Phytologist, Vol.XI, 1912.
89. Wettstein, R. R. V. Handbuch der Systematischen Botanik. Leipzig, 1901.
RHIZOPHIDIUM ACUFORME (ZOPF) FISCH.
BY W . B. GROVE, M.A.
[WITH ONE FIGURE IN THE TEXT].
ON Chlamydomonas intermedia, of which a large gathering wasmade at Harborne (in the same cart-rut that is mentioned
in the following note), there was found in April, 1917, a quantityof Rhizophidium acuforme. This parasite has been previouslyrecorded by Zopf on a Chlamydomonas-Vike organism in Marchfrom an ice-covered pond in Pomerania, but there remained adoubt how far it was really distinct from 7?. globosum (Braun),which occurs on a large number of fresh-water Algse. This doubtis dispelled by the present find.
The cells of the Chlamydomonas are attacked by the zoosporesof the Rhizophidium, while still in active motion, at any pointof their circumference ; one to four (or even five or six) may he
