■^^^'iT^tiL

oQc

tfVW

:WM<^,

w*

Km

If 1

■**r^*

.»:i_t_i.i

I ».i?

» » *

iU^rn'f .iLt

!».»

^.^^.-^.i "'.^ '^ ^:' -^

iim#:f "' #-,#^t-i- *•»-»- t,tw*

«• f

t fi K

!.*.»

»'%■

? !• »' t

*■' *

f#

.*,

E***t*' *■■*-*♦ *'^-*'*'

« « 1 , .. f ^ ^

5, * >.t.,f ..»„

.♦,*.. «u*.

i*f ■* H». iiik,

^mmm

mmmmmmmMmmm

1

HARVARD UNIVERSITY

Library of the

Museum of

Comparative Zoology

BULLETIN

OP THE

MUSEUM OF COMPARATIVE ZOOLOGY

AT

HARVARD COLLEGE, IN CAMBRIDGE

VOL. 115

CAMBRIDGE, MASS., U.S.A. 1956-1957

The Cosmos Press, Inc. Cambridge, Mass., U.S.A.

.0 0 I

CONTENTS

PAGE

No. 1. Notes on American Earthworms of the Family LuMBBiciDAE. ITI-VII. By G. E. Gates. August, 1956 1

No. 2. A Quantitative Study of the Equidae of the Thomas Farm Miocene. By Robert S. Bader. Au- gust, 1956 . . . ^ 47

No. 3. Aneuretiis simoni Emery, a Major Link in Ant Evolution. By E. 0. Wilson, T. Eisner, G. C. Wheeler and J. Wheeler. (3 plates.) August, 1956 79

No. 4. The Types of Naiades (Mollusca: Unionidae) IN the Museum of Comparative Zoology. By Richard I. Johnson. September, 1956 ... 99

No. 5. Pseudemys scripta callirostris from Vene- zuela w^iTH A General Survey of the scripta Series. By Ernest Williams. (3 plates.) Septem- ber, 1956 143

No. 6. Revision of the African Tortoises and Tltitles of the suborder Cryptodira. By Arthur Lov- eridge and Ernest E. Williams. (18 plates.) Feb- ruary, 1957 161

Bulletin of the Museum of Comparative Zoology

AT HARVARD COLLEGE Vol. 115, No. 1

NOTES ON AMERICAN EARTHWORMS OF THE FAMILY LUMBRICIDAE. IIl-VII.

By G. E. Gates

CAMBRIDGE, MASS., U. S. A.

PRINTED FOR THE MUSEUM

August, 1956

Publications Issued by or in Connection

WITH THE

MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE

Bulletin (octavo) 1863 The current volume is Vol. 115.

Breviora (octavo) 1952 No. 57 is current.

Memoirs (quarto) 1864-1938 Publication was terminated with Vol. 55.

Johnsonia (quarto) 1941 A publication of the Department of MoUusks. Vol. 3, no. 35 is current.

Occasional Papers of the Department of Mollusks (octavo) 1945 Vol. 1, no. 18 is current.

Proceedings of the New England Zoological Club (octavo) 1899- 1948 Published in connection with the Museum. Publication terminated with Vol. 24.

The continuing publications are issued at irregular intervals in numbers which may be purchased separately. Prices and lists may be obtained on application to the Director of the Museum of Comparative Zoology, Cambridge 38, Massachusetts.

Of the Peters "Check List of Birds of the World," volumes 1-3 are out of print; volumes 4 and 6 may be obtained from the Harvard University Press; volumes 5 and 7 are sold by the Museum, and future volumes will be published under Museum auspices.

Bulletin of the Museum of Compqrative Zoology

AT HARVARD COLLEGE Vol. 115, No. 1

NOTES ON AMERICAN EARTHWORMS OF THE FAMILY LUMBRICIDAE. III-VII.

By G. E. Gates

CAMBRIDGE, MASS., U. S. A. PRINTED FOR THE MUSEUM

August, 1956

No. 1 Notes on American earthworms of the family Lumbricidae. III-VII.

By G. E. Gates III

BiMASTOs TUMiDus (Eiseii, 1874)

Allolobophora tumida Eisen 1874, Ofv. Vet. Ak. Forh., 31, p. 45. (Type

locality, Mt. Lelnuion, New York-Massachusetts. Type, one half of

anterior end sagittally sectioned by Smith, in U. S. Nat. Mus.) Allolohophora gic.srlcri Ude 1895, Zool. Aiiz., IS, p. 339, Zeit. wiss. Zool.,

61, p. 127. (Type locality, Savannah, Georgia. Types, numerous, in

Hanover Provincial Mus.?) Allolohophora (Bimastm) tumida + -4. (-B.) gieseleri, Hclodrilus (B.) t.

+ E. (B.) g., Michaelsen, 1900, Abh. Nat. Ver. Hamburg, 16, p. 16,

Das Tierreich, 10, p. 502. Helodrilu^ gieseleri hempeli Smith 1915, Bull. Illinois Sta. Lab. Nat. Hist.,

10, p. 551. (Type locality, Havana, Illinois. Tj-pes in U. S. Nat. Mus.) Helodrilus (Bimastus) tumidus + S. (B.) gieseleri f. tijpica + H. (B.)

gieseleri var. hempeli, Smith, 1917, Proc. U. S. Nat. Mus., 52, pp. 170-

172.

? Bimastus duels Stephenson 1933, Proc. Zool. Soc. London, 1932, p, 939.

(Type locality, Durham, North Carolina. Holotj^pe in Brit. Mus.)

Tamalpais, Marin County, California, 0-0-1 (macerated). California Ac.

Sci. New Orleans, Louisiana. Under logs in moist river bottom forest, Dec.

31-Jan. 1, 1931-32, 0-10. J. M. Valentine per Dr. G. E. Pickford. Big Thicket, Texas, May 1, 1938, 0-0-1. Ottys Sanders. Athens, Texas. Six miles to the north, April 3, 1948, 0-0-3. Ottys Sanders. Paris, Texas. Eleven miles to the south, March 29, 1952, 0-0 1. Ottys

Sanders. (Three specimens, without label, from Mr. Sanders probably

also from Texas.) North Dublin, Virginia. Sawdust, Little Walker Creek, June 25, 1955,

0-2-17. Prof. Walter Harman. South Rushton, Louisiana. In and near ash of sawdust pile, April 2, 1955,

0-0-3. Prof. Walter Harman. Lincoln Parish, Louisiana, March 10, 1955, 0-0-2. Bobby Wise per Prof.

Walter Harman.

2 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

External characteristics. Length, 26-40 mm. Diameter, 2-3.5 mm. Segments, 81-122 (ef. Table 1). Anal segment of worms not recently amputated usnally longer than penultimate one and Avith setae in middle of anterior half but with little or no indica- tion of a dorsal pore or an intersegmental furrow that would mark off setigerous portion into a metamere. Body approximat- ing to four-sided posteriorly, the dorsum slightly rounded and wider than the ventrum, one setal couple at each corner. Pig- mentation (unrecognizable in alcoholics except at site of clitel- lum in a Louisiana worm) red, in dorsum, reaching ventrally to or nearly to B through most of axis, also present in ventrum of some or all of first fourteen segments, lacking at intersegmental furrows, in wide transverse stripes in clitellar segments (recently preserved Virginia worms). Prostomium epilobous, tongue open (all). Setae begin on ii on which all usually are present, paired but not as closely as in some lumbricids, d about at mL anteriorly but more dorsal posteriorly where CD a t rifle <AB, BC<.AA, DD<^iC hn\y AA. Nephropores, recognizable only on a few segments, just above B on xv (all worms on which pores of that segment were recognizable), on clitellum irregularly and some- times asymmetrically alternated between levels just above B and well above D (Virginia). Possible sites on more anterior seg- ments have been noted on several specimens after some softening and removal of the cuticle ; at or above D on iii-vi, slightly above B on vii-ix, above D on xi. First dorsal pore on 5/6 (all).

Female and male pores equatorial and at usual levels on xiv and XV respectively. Male pore tumescences in median half of BC, confined to xv but sometimes dislocating 14/15 and 15/16 slightly, occasionally reaching on one or both sides slightlj^ into xvi.

Clitellum saddle-shaped, reaching ventrally nearly to B (Louisiana) or to A, located as follows; xx-xxix (1, Louisiana), xx-xxx (2, Texas), (xx-xxi?) xxii-xxix (2, Texas), (xxi?) xxii- xxix (1, Texas), posterior part/xxi-anterior part/xxx (1, Vir- ginia), (xxi?) xxii-xxx (1, Louisiana), xxii-xxix (14, Virginia and 4, Louisiana), xxii-xxix and onto xxx (2, Virginia), xxii- xxix and possibly xxx (2, Texas), xxii-xxx (3, Virginia, Texas and California). Tubercula pubertatis lacking.

Genital tumescences lacking or unreeognizable.

GATES : AMERICAN LUMBRICIDAE 3

Internal anatomy. No septa thickly muscular. Longitudinal muscle band at niD with pigment, scattered flecks of red pigment present on (in or under?) peritoneum of dorsum anteriorly (10, Virginia).

Calciferous gland extends well into x, the posterior portion of the gut in that segment often with an appearance as of a pair of vertical sacs. Gland of about the same calibre through xi-xii, nneonstricted at insertion of 11/12. Intestinal origin in xv (24). Gizzard in xvii-xviii (24). Typhlosole begins gradually in leaion of xxi and in places may reach floor of gut, slightly widened veutrally in an anterior portion and there with a single median groove. The typhlosole ends abruptly as shown in the table.

Subneural trunk unrecognizable (22), very narrow but in part red and on ventral face of cord (2). Extra-oesophageal trunks empty, apparently passing into dorsal trunk posteriorly in xii (1, Virginia). Hearts present in vii-xi, those of vii empty in two worms (Virginia). Nephridial ducts, at least behind xv, pass into parietes at B. Vesicle, U-shaped.

Table 1

Typhlosole termination and segment number

in Bimasios tumidus

Typhlosole

Atyphlo-

ends in

solate

Number of

segment

segments

segments

Remarks

81

19

100

Posterior amputee

83

17

91

*

Posterior amputee ?

85

15

100

Posterior amputee ?

95

20

115

97

16

113

« .1

Fcisteiiur aiiiputees?

97

20

117

Louisiana

98

19

117

Posterior amputee ?

99

12

111

»

Posterior amputee ?

101

13

114

*

Posterior amputee ?

103

15

118

Louisiana

106

16

122

Louisiana

107

15

122

Louisiana

Last segment with setae as indicated on page 1. AU worms from Virs'inia cxceiit as otherwise indicated.

Segment numher of other wnims : St (posterior amputee). 107 (posterior amputee), ins, ii:;. 120.

4 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

Seminal vesicles rather small, two pairs, in xi-xii (24). Male funnels crenellate (24). Male deferent ducts slender, shortly looped just behind funnel septa, the loops crowded into a small, leaf-shaped disc. The ducts of a side come into contact in xii where they apparently unite. They disappear from sight at anterior margin of atrial gland or are recognizable further pos- teriorly on surface of the gland (24). The ducts must have grown posteriorly to or nearly to eq/xv, in some of these worms, before the developing atrial gland had penetrated through the longitudinal musculature. Spermathecae lacking (24). Ovisacs of about the same size as the ovaries and slightly lobed (24).

Atrial gland crossed in xv by an equatorial cleft, a second cleft possibly at insertion of 15/16 as the gland seems to extend slightly into xvi. No glands on parietes in BC of clitellar seg- ments nor in association with setal follicles. No setal follicles especially protuberant into coelom (no genital setae?) in four- teen Virginia and Louisiana worms.

Abnormality. Spiral metamerism involving v-vi only (1, Virginia). Metameric abnormality in preanal region (1, Vir- ginia).

Life history. Slight iridescence was recognizable on each male funnel in ten Virginia worms collected on the 25th and 27th of June, as well as in four from Louisiana (March 10 and April 2). Male deferent ducts also appeared to be slightly iridescent. No spermatophores. Seminal vesicles of an aclitellate Virginia worm were of the same small size as in clitellate specimens and had small brown bodies near dorsal ends, but none of the usual ex- ternal indications of postreproductive regression was recognized. The clitellum of some Virginia and five Louisiana worms may have been at maximal tumescence. Sparse sperm production ap- parently is indicated by these worms, in which case uniparental reproduction perhaps should be anticipated, possibly' partheno- genesis.

Parasites. Cysts, in coelomic cavities of anterior half of body of a Texas worm, are attached to parietes and septa by long filaments.

Remarks. Condition of the author's worms (most of them alcoholic) is fair to very poor; the softened body wall of the California worm was alreadv broken through before removal

GATES : AMERICAN LU:mbRICIDAE 5

from the tube in which it arrived. Much better material was, however, loaned by Prof. Ilarnian, a favor which is much ap- preciated.

Spermathecal vestijres were not recognizable in the parietes after removal of lon<i'itudinal musculature but broad transverse bands of pigmentation became quite obvious.

Septa 6/7 or 7/8 to 11/12 or 14/15 are somewhat strengthened but none is thickly muscular. The thickness seems to vary with the contraction or relaxation of the worm at preservation as well as with degree of maceration. Differences detectable in dis- section, insofar as present material is concerned, are slight and do not appear to be of any taxonomic value.

Function and phylogeny of the gland present on the parietes over site of male pore and its tumescences in many lumbricids are unknown. The characterization "atrial" is retained, for want of a bettor term, but without any connotations as to func- tion or evolutionary history.

The clitellum of one Texas worm was about as in tiimidus, of two as in gieseleri, of another as in hempeJi. The dift'erences ap- pear to be no greater than are now to be expected in normal intra- specific variation.

B. duels, known only from the original description of the type, was thought to be distinguished, presumably from tumidus and gieseleri, by a longer clitellum of 11^ segments. The last hearts of duds are in x, according to Stephenson, but hearts of xi when empty may not be readily recognizable in ordinary museum material. Number of chambers in the calciferous gland, which Smith (1917) believed to be of taxonomic importance (though he did not always mention them) is unknown.

The clitellum covers 8-11 segments in tumidus and another half segment may be within the range of normal intraspecific variation in Bimastos. Location of anterior boundary of the clitellum, at or in front of 21/22, also characterizes tiimidus but no other species of Bimastos. Coverage of xxxi-xxxii by the cli- tellum now seems to be the only reason that can be cited in sup- port of retention of duels.

B. longicinctus Smith and Gittens 1915 has a clitellum, in Illinois, according to its authors, of 10-11 segments, on xxiii-xxxii or xxxiii. Location in worms subsequently recorded by others from Ohio and Arkansas without data as to variation presumably

6 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

was the same, as the organ is all that enables routine identifica- tion of the species. Number of segments covered is about the same as in tumidus, but the anterior margin is at 22/23 rather than at 21/22 or anteriorly. The only other method available for separating these two forms involves counting chambers or partitions in the calciferous gland. There are about 60 of these in longicinctus, about 40 in tumidus. The difference, according to Smith (1917, p. 175), is sufficient to warrant specific distinction. Although little is known as yet about variation in such a charac- ter, requiring microtome sections, the number of partitions in the calciferous gland of Eisenia rosea (Savigny, 1826) is 38-54.

IV

BiMASTOS PARVUS (Eisen, 1874)

Allolobophora parva Eisen 1874, Ofv. Vet. Ak. Forh., 31, p. 36. (Type

locality, Mt. Lebanon, New York-Massachusetts. Types in TJ. S. Nat.

Mus.) Allolobopliora heddardi Michaelsen 1894, Zool. Jahrb. Syst., 8, p. 182.

(Type locality, Orlando, Florida. Type in Hamburg Mus.?) Allolobophora (Bimastus) parva + A. (B.) beddardi, Hclodrilus (B.) p.

-f H. (B.) h., Michaelsen, 1900, Abb. Nat. Ver. Hamburg, 16, pp. 14-13,

Das Tierreieh, 10, p. 502. Eelodrilus {Bimastus) parvm -f H. (B.) beddardi, Smith, 1917, Proe.

U. S. Nat. Mus., 52, p. 173. Bimastus parvus -f B. beddardi, KobayasM, 1940, Sci. Eep. Tohoku Univ.,

15, pp. 297, 298. Bimastus beddardi, Cemosvitov and Evans, 1947, Linnean Soc. London,

Synopses British Fauna, 6, p. 18.

Mexico. In soil around roots of plants from Guadalajara, Jalisco, in baggage arriving at Nogales, May 10, 1949, 0-0-1.

Fish bait in soil from Juarez, Chihuahua, in passenger's baggage arriving at El Paso, May 21, 1949, 0-3-2.

In soil with two plants in passenger's baggage arriving at Santa Fe bridge, August 13, 1953, 0-0-1.

In soil with plants from Torreon, Coahuila, in passenger's baggage arriving at Santa Fe bridge, August 13, 1953, 0-0-1. Italy. In soil witli strawberry and rosemary plants in baggage on SS

Atlantic arriving at New York, March 22, 1950, 0-0-1. .lapan. In 25 ll)s. soil with plants in baggage on SS Cavnlier ainivin.? at San Pedro, April 23, 1954, 0 0-1.

GATES: AMERICAN LUMBRICIDAE 7

England. In soil with plants in baggage on SS Britannic arriving at

New York, April 17, 1949, 0-1-1. Wales. In soil with plants in baggage on SS Britannic arriving at New York, April 24, 1949, 0-0-3.

(Worms listed above were intercepted by officers of the Division of FoT-oign Plant Quarantines, U. S. Department of Agriculture and were available through courtesy of Dr. C. W. F. Muesebeck and Dr. P. W. Oman.) Texas (probably), 0-0-3. Ottys Sanders. (In an unlabelled tube along with B. tumidus.)

External eharacterisfics. Length, 25-42 mm. Diameter, at clitellum, 2-3 mm. Segments, 101 (1), 102 (1), 104 (1), 109 (2). Pigmentation, whenever recognizal)le (formalin preservation), red. Body rather four-sided posteriorly. Prostomium epiloboiis, tongue open. Nephropores (-when recognizable) just above B or above D. First dorsal pore at 5/6.

Female and male pores e(iuatorial. Male pore tumescences confined to xv or extending well into xiv and xvi (10).

Clitellum saddle-shaped, on (xxiii?) xxiv-xxx (1), (xxiii?) .\xiv-xxxi (2), (xxiii?) xxiv-xxxii (2), xxiv-xxx (2), xxiv-xxxii (1), xxiv/2-xxxii/2 (3), xxv-xxxi (9). Tubercula pubertatis lacking or possibly represented by quite indistinct areas on xxvi- .Kxx (several).

Internal afiatomy. Calciferous gland, intestinal origin, gizzard and typhlosole as in tuniiclus. The typhlosole has no ventral groove anteriorly and, in a worm of 109 segments, ends in xcvi. Hearts were recognizable only in vii-xi (1) or viii-xi.

Seminal vesicles quite small. Atrial glands markedly pro- tuberant into coelom. No setal follicles especially protuberant from parietes (genital setae lacking?).

Life Jiistory. Spermatozoal iridescence was not recognized on male funnels and spermatophores were lacking externally though the clitellum in some of the worms appeared to be at maximal tumescence. Parthenogenesis and male sterility are sus- ]:)ected.

Abnormality and variation. Helicometameres were noted be- hind clitellar region in two worms. Seminal vesicles were lack- ing in ix but were present in x of two worms.

Remarks. Many worms lacked the posterior end and most were poorly preserved. If pigment once was present in alcoholic

8 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY

material it had become unrecognizable. Anterior and posterior margins of clitellum, as in tumidus, occasionally may not be distinct because of quite gradual decrease in height of epidermis. Whether male deferent ducts are looped on posterior faces of funnel septa Avas not determinable.

If the usual assumption as to American origin of B. parvus is correct, this species is the only American lumbricid that has become "peregrine". It had been recorded, usually without data as to variation, from at least sixteen states, as well as from Mexico, Guatemala, Brazil, Hawaii, Japan, Manchuria, China. Tibet, India, Burma, the Malay Peninsula, Java, Germany and the British Isles.

Two species recognized by Smith (1917), B. parvus and hed- dardi, were distinguished from each other almost only by dif- ferences in clitellar location and in intersetal spacings- that now appear to be within limits of normal intraspecific variation.

B. parvus is distinguishable from tumidus, also with about 40 partitions or chambers in the calciferous gland, only by clitellar characteristics. The organ tends to be shorter, usually of 7-8 segments, though 9 have been found and 10 may be possible. Situation of anterior margin at 23/24 or 24/25, even at 22/23 when anterior margin is less definite, provides a further distinction but not, in the latter case, from longicinctus. To that species one such specimen, with a clitellum also covering xxxi- xxxii, would have been referred but for its foreign origin.

So much emphasis on the clitellum is required because of more or less recent elimination in Bimastos of spermathecae, tubercula pubertatis, perhaps also genital setae and certain glands. Species usually differ most obviously from one another, in the Lumbrici- dae as well as other earthworm families, in their reproductive organs on which the classical taxonomy largely has been based. Satisfactory substitutes for the characters that are no longer available still have to be found, not only in the Lumbricidae but also in other genera where similar evolutionary changes have been or still are under way (cf. Gates, 1954, p. 237).

GATES: AMERICAN LUMBBICIDAE 9

ADDENDUM

EisENiA ROSEA (Savigiiy, 1826)

External characteristics, of several American alcoholic speci- mens, are such as to permit identification as B. parvus. Just as in that species, seminal vesicles of ix-x and all spermathecae are lacking. However, enlarged setal follicles are protuberant, in each worm, into coelomic cavities of one or more segments ; a-h or c-f//x-xiii, a-b/xxx, xxxii. Such follicles, with genital setae, are to be expected in those very segments in the exotic rosea where their presence usually is indicated by genital tumescences, but apparently not in any American species of Bimastos.

The calciferous glands of rosea were described by Smith (1924) immediately after those of parvus. The only difference mentioned is in width of lamellae and of chambers.

Tubercula pubertatis have been lost during evolution of an A morph ( ''Bimastos stage") in species of several lumbricid genera, and their absence in these athecal individuals of rosea is not, then, surprising.

Follicles that are to produce genital setae become enlarged, in this as well as other exotic species, during a juvenile stage before development of tubercula and clitellum which appear ontogenetically in that order. Tumescences develop around aper- tures of GS follicles as the clitellum is differentiated. Suppres- sion of GS tumescences in this American parthenogenetic strain may be an early stage in an evolutionary development that will result in elimination of the associated GS follicles and their modified setae. Just such an early stage apparently has been reached by certain species of Bimastos. In others, tumidus and parvus, such evolutionary development may have been com- pleted. Genital setae, presumably functioning only during copulation, are of no use to animals that have become partheno- genetic.

Bimastos palustris Moore 1895

Xew Jersey, 0-0-7. Harold Davies (1954). External characteristics. Length, 18-38 mm. Diameter, 1 mm. or slightly more at posterior end, ca. 2 mm. anteriorly, 2+ mm. in clitellum, 3 mm. in xv-xvi. Segments, to 101 (cf. Table 2).

10 BULLETIN : MUSEUM OP COMPARATIVE ZOOLOGY

Pigmentation red, in circular musculature of dorsum anteriorly, lacking close to intersegmental furrows but present in ventrum of first five or six segments, recognizable only 'after stripping off longitudinal musculature, a light pink coloration all that is vis- ible externally. Body squarish posteriorly and with setae at corners. Prostomial tongue broad, short, open (7). Setae begin on ii on which all are present (7), AB=, < or > CD, but dif- ference slight, BC < AA < DD < IC. Nephropores certainly recognizable on'y behind xv, irregularly and asymmetrically alternate ])etween levels just above B and dorsal to D. First dorsal pore on 5/6 (6), W/7 (1), a weak spot at mD on 4/5 in one worm clearlj^ not a definite pore.

Female pores diagonal to transverse slits (or in such slits?), posterolateral to h (7), at most midwaj^ between eq/xv and 15/16. Male pores deeply invaginate, the usual transverse cleft apparently slightly postequatorial in some but probably really equatorial on all. Male pore tumescences conspicuous, maximal height at or close to 16/17, sloping anteriorly down to the cleft, scarcely noticeable anterior to eq/xv.

Clitellum annular (7), epidermis not as thick in AA as dor- sally but intersegmental furrows unrecognizable (6), anterior and posterior margins distinct and as much so ventrally as in the dorsum (7), on xxiii-xxvii (1), xxiii-xxviii (6). No tubercula pubertatis (7).

Genital tumescences include a-h setae, on xiii (7) and xvi (7).

Internal anatomij. Septa l:)aek to 14/15 somewhat more opaque than posteriorly, 8/9 apparently the thickest, 7/8-9/10 less so, all weak merely pushing gut over to one side or the other breaks the septa at parietal insertions, even anteriorly. Pigment apparently lacking in longitudinal muscle band at mD and in ]ieritoneum.

Calciferous gland extends into x (7), the gut in that segmen*: with an appearance as of two vertical sacs posteriorly. Constric- tions at 10/11-11/12 usually slight or unrecognizable but in one specimen so deep externally that the gland appears to be monili- form as in Eisenia Idnnbergi (Michaelsen, 1894) and F. carolin- ensis Michaelsen 1910. Constriction at or near 12/13 external or internal or both and then different on opposite sides. Intestinal origin in xv (7). Gizzard in xvii-xviii (7). Typhloscle begins rather gradually behind the gizzard and is a simpli- lamella,

GATES : AMERICAN LUMBRICIDAE

11

slijihtlj' thicker ventrally, often reaching fioor of gut, ending as shown in Table 2.

Snbncural trunk mostly unrecognizable but occasionally indi- cated, on ventral face of cord, in two worms by an indistinct red streak visible only in a segment or two. Extra-oesophageals, pre- sumai)ly empty, unrecognizable. Hearts in vii-xi (7), those of \ii em])ty and small (4), all empty (1). A vertical red column, resembling a blood vessel and doubtless of blood, recognizable in lateral Avail of calciferous gland on each side in one or more or all of x-xii. Nephridial ducts, at least beliind xv, pass into j)arietes at B.

Tabi>k 2

Typldosole termination and segment number in Bimasfos lyahtsiris aiul B. sp.

Typhlosole

Atyphlo-

Number

ends in

solate

of

segnient

segments

segments

Remarks

47

8

55

Posterior amputee

48

4

52

Posterior amputee

62

6

68

Posterior amputee

69

8

77

Posterior amputee

83

10

93

Posterior amputee?

84

11

95

Posterior amputee?

87

14

101

B.

sp.

101+ ?

?

101

Posterior portion broken oiT and lacking.

130

17

147

Seminal vesicles equisized and fairly large, in xi-xii (7). Male funnels large, crenellate, reaching well up at sides of gut. Male deferent ducts straight, those of a side coming into contact in xii, passing up to apex of atrium where primary male pore doubtless is located. Spermathecae lacking, rudiments unrecog- nizable in parietes after removal of longitudinal musculature (7). Ovisacs present (7). lobed, smaller than the ovaries. 0\\-

12 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY

ducts seem to be unusually thick relative to size of worm and obviously pass into parietes behind eq/xiv.

Atrium thumb-shaped, erect in xv, reaching well up toward level of dorsal face of gut (7), basal portion distended pos- teriorly and continued into xvi as a transversely placed reni- form mass that bulges back 16/17 at parietes. No corresponding extension on parietes anteriorly.

Follicles of ventral setae of xiii and xvi slightly thickened but thin- walled, elongated, apices reaching into coelomic cavities just above associated glands. Acinous glandular masses on parietes median and posterior to each such pair of follicles.

Ah7iorniality. Male pore of right side and atrium in xvi, right male tumescence in xvi-xvii. Basal gland of atrium in xvii. Geni- tal tumescences of right side in xiii and xvii. Clitellum not as thick ventrally in BB as in other worms and intersegmental furrows faintly indicated across mV.

Habitats. Apparently restricted to banks that are kept moist by running water (cf. Moore, 1895 and Davies, 1954).

Natural history. Slight iridescence is recognizable on male funnels of two specimens with spermatophores, but not on funnels of other worms.

Spermatophores were found by Moore from latter part of February to December. Cocoons may then be deposited in any season of the year that temperature permits. Although not now anticipated in the summer, some sort of a diapause may be expected in January or whenever the weather is too cold for normal activity.

Biparental reproduction can be assumed, at present, as Moore mentioned release from spermatophores of sperm presumably normal.

Distrihtition. Pennsylvania, within thirty miles of Phila- delphia. Northern New Jersey. Raleigh, North Carolina. The latter record was accompanied by no information as to habitat, number of specimens, structure, etc.

Remarks. Greater opacity of epidermis than usual in the dorsum anteriorly, in vivo as well as after preservation, pre- sumably is responsible for a previous assumption as to absence of pigment (cf. Smith, 1917, p. 169).

Spermatophores are 2-2-|- mm. long, attached to xiv in DD.

Male pore tumescences in the types, so far as can be judged

GATES: AMERICAN LUMBRICIDAE 13

from the figure provided by Moore, seem to have been as marked anterior to the cleft as behind it.

The calc'iferous gland appears to be constricted at or near insertion of 12/13 as in various other American lumbricids both exotic and endemic. The constriction may be external on one side and internal on the other or internal all around when gut is straiglit. These dift'erences probably are of little significance, perhaps a matter of differential contraction at time of preserva- tion. However, the constriction is usual and in dissections ap- pears to provide a posterior boundary of the gland though calcif- erous lanudlae, according to Smith (1924) frequently extend into xiii.

Hearts tiid not pass in any of these specimens into gut wall ((•r. Moore, 1895) which confirms Smith's finding (1928, p. 354). The vei'tical cords of blood mentioned above may have been mistaken for hearts. Inability to trace the extra-oesophageal trunks in the present specimens is unfortunate because of a possibility that they may provide characters of taxonomic im- portance in a group where such are much needed. However, it can be suggested, after examination of the vessels in many speci- mens of various other species that the extra-oesophageal trunks do not join the ventral trunk in x (Moore, 1895) or the dorsal trunk in x (Smith, 1928). Distention by blood of an anterior portion of the extra-oesophageal and of a fairly large branch that occasionally passes (in other species) to the dorsal trunk in ix and/or x, with the portion of the extra-oesophageal behind the branch empty and collapsed, may have been the condition seen by Smith. A branch from the subneural arising in any of segments xiv-x (again in other species) may pass to the extra- oesophageal. AYhen that branch is distended with blood the trunk at first appears to turn laterally and join the extra-oeso- phageal but does again become recognizable anteriorly wherever blood is present. The diificulties posed by temporarily empty vessels in earthworms rarely have been appreciated even by those who made special studies of the circulatory systems.

Microscopic examination of the setae in the enlarged ventral follicles of xiii and xvi has not been possible. Some modification certainly can be expected even though not recognized by Moore nor mentioned by Smith.

14 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY

B. palustris is distinguished from the previously erected species, tumid'us and parvus, by the shorter clitellum of only 5-6 segments as well as its obvious annularity, posterior disloca- tion of female pores, deep invagination into coelom of male pores at apices of large atria, the enlarged setal follicles in xiii and xvi, the epidermal tumescences and the coelomic glands ("postsetal", Moore) associated with those follicles. The folli- cles, genital setae therein, glands and tumescences have not been used in the classical taxonomy. Little is known, however, about the variation of any of these characters in this species or of the newer ones just mentioned in any lumbricid. Another dis- tinction from tumidus and parvus may be in the method of reproduction.

Closest to palustris of the Bimastos species now appears to be heimburgeri Smith 1928, erected on a single specimen from Illinois. Four individuals subsequently recorded from North Carolina, in absence of any description, must be assumed to be like the type. The species is distinguished from palustris by the clitellum, of 8-|- segments on xxv-xxxiii(part) and saddle- shaped, absence of genital setae and presumably also of the aci- nous glands. The type was not "sexual" though clitellate, and Smith suggested (1924, p. 856) that genital setae "may be present at the time of sexual activity." This now seems unlikely. Enlargement and protuberance into coelom of GS follicles is obvious in lumbricids studied by the author before any indica- tion of clitellar development is recognizable, and in thecal species even before appearance of the tubereula pubertatis. If Smith's statement re gonads of diminished size referred to the testes, and if clitellar tumescence was maximal, the type may be par- thenogenetic. Otherwise iridescence should have been recogniz- able on the male funnels of the unsectioned half and in the sec- tions of the other half sperm on the funnels should have been obvious.

Bimastos sp.

Moonshine Dell, Giles County, Virginia, June 20, 1955, 0-0 2. Prof. Waltpr Plarman. External characteristics. Size, 71: x 4.5 (at xv) mm. Seg- ments, 147. Pigmentation red, in dorsum but sparse behind clitellum until near hind end, in ventrum of first seven seg-

GATES: AMERICAN LUMBRICIDAE 15

merits. Body approximating to transversely rectangular in cross section posteriorly, M-itli setae at the corners. Setae begin on ii on which all are present, rather closely paired, AB ca. = CD, BC ca. = AA, DD < iC posteriorly. Prostomium epilobous, tongue open (2 specimens). First dorsal pore on 5/6 (2). Nephropores unrecognizable.

Clitellum annular, only slightly tumescent, intersegmental furrows recognizable all around, on xxiii-xxxi (1), still quite indistinct on the other but apparently indicated by a whitening of the dorsum on xxv-xxx. Tubercula pubertatis lacking.

Female pores definitely equatorial on xiv, of about the same size and shape as adjacent aperture of a GS follicle. Male pores each in a deep equatorial cleft or slit at bottom of a depression sloping down from 15/16 and 14/15, tumescences markedly pro- tuberant, extendino- across all or nearly all of xiv and xvi, equally developed anterior and posterior to eq/xv.

Genital tumescences unrecognizable (perhaps not yet de- veloped) but a-h setae of xii-xviii retracted into parietes, follicle apertures enlarged, slit- or comma-shaped, longitudinally placed.

hiternal anatomy. Septa 8/9-14/15 strengthened, 12/13 the thickest. Pigment sparse in longitudinal muscle band at mD. Peritoneal flecks very few.

The calciferous gland extends well into x and there is widened so that there is an appearance as of two vertical sacs. A slight constriction at insertion of 10/11, none at 11/12, internal at 12/13. Intestinal origin in xv (2). Gizzard in xvii-xviii (2). Typhlosole begins in xxii, nearly 2 mm. high but rather thin anteriorh- where it reaches floor of gut, and has a deep longi- tudinal groove on ventral face, ending abruptly (cf. Table 2).

Subneural trunk recognizable only for a distance of one seg- ment here and there where a little blood is present. Extra-oeso- phageals empty but traceable (1) into dorsal trunk posteriorly in xii. Hearts present in vii-xi, empty in vii (1) or with a little blood but small. Nephridial ducts pass into parietes at B.

Seminal vesicles rather small, tough, in xi-xii. Male funnels crenellate. Male deferent dacts without looping on funnel septa. Ovaries and ovisacs as usual, spermathecae absent (2).

Atrium hemispheroiclal, the lumen reaching above level of parietes, with glandular extensions into xvi and xiv. Follicles of ventral setae of xii-xviii enlarged and conspicuously pro-

16 BI^LLETIN : MUSEUM OP COJNTPARATn'E ZOOLOGY

tuberant above parietes. Acinous glands apparently are associ- ated with the follicles.

Remarks. Thickened septa, nephridia, GS follicles and acinous glands are closely crowded and adherent to each other. Deter- mination of relationships in such circumstances sometimes be- comes possible after several months further treatment with preservative.

Enlarged apertures of GS follicles presumably are common external openings of the follicles and of the associated acinous glands.

Atria are not as high as in palustris.

Spermatozoal iridescence on male funnels is slight. Such iridescence has been noted in other lumbricid species during postsexual regression, at onset of maturity, and when matura- tion is sparse as in certain supposedly parthenogenetic forms.

These worms are to be compared with those of species having the anterior margin of the clitellum at 22/23, palustris, longi- cinctus and parvus (only occasionally). They cannot go in palustris as at present understood though they do have GS follicles in xiii and xvi and acinous glands associated therewith. Presence of those follicles and glands presumably rules out 2)arvi(s in which they are lacking. Nothing is known as to GS follicles and glands in longicinctiis in which the clitellum is sad- dle-shaped though this latter difference may prove to be of no importance.

BiMASTOs sp.

Whitetop Mountain, Viiginia, under bark of log at elevation of 4700 feet, August 17, 1955, 0-0-1. Prof. Walter Harman.

External characteristics. Size, 16 x 2 mm. (through clitellum). Segments, 61 (probable posterior amputee). Pigmentation red, obvious externally. First dorsal pore at 5/6.

Female pores on xiv, further from B than usual, almost at level of male pore cleft, only slightly if at all postequatorial. Male pore clefts equatorial on xv. Male tumescences extend well into xvi and xiv, 1-1/15 and 15/16 unrecognizable across them, and include a-h of xv-xvi. Clitellum annular, on xxiv- xxxii(xxxiii?), posterior margin indistinct. No tubercula.

Internal anatomy. As in previous species except as otherwise mentioned.

GATES: AMERICAN LUMBRICIDAE 17

Typhlosole ends in liii but is much lower there than in the preceding segment.

Male deferent ducts looped between funnel septa and parietes but in a loosely zigzagged manner, not compacted into a ball or disc. Oviducts pass into parietes anterior to equator of xiv and are relatively thick. GS follicles, a-b of xv-xvi, markedly pro- tuberant into coelomic cavities. No acinous glands associated with those follicles.

No atria ; even atrial glands lacking.

Parasites. Five transparent cysts, each of about the same size as the prostomium, are present in coelomic cavities of xi-xii.

Eemarks. No iridescence recognizable on male funnels.

Species with the anterior margin of the clitellum at 23/24 are longicinctus and parvus. Presence of GS follicles in xv-xvi (that are lacking in parvus) apparently obviates further consideration of that species. Nothing is known as to presence or absence of atrial glands in longicinctus.

VI.

Species of Eisenia

The distinctness and relationships of our endemic species of Eisenia, which were characterized about as well as their Eu- ropean congeners, have been misunderstood abroad. So little has been known about variation in and distribution of the American forms that the collections made by Dr. Pickford and Prof. Har- man, in the hitherto uninvestigated mountains of North Caro- lina, Tennessee and Virginia (Gates, 1955), though secured in a probably unfavorable season, have joroved to be of considerable importance. Material received since completion of the manu- script of parts I-II of this series now has provided some of the additional information that was w^anted for a discussion of these relationships.

E. hortensis obviously is exotic in North America. A brief description is included to provide hitherto unrecorded but now necessary information. Notes on other exotics, E. foetida and rosea, are provided for similar reasons.

The author's thanks are extended to Prof. Walter Harman, Harold Davies, and Curator Fenner Chace, Jr., for the oppor- tunity of studj-ing the various specimens they supplied.

18 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY

Genus ElSENIA Malm 1877

EiSENiA CAROLiNENSis Michaelsen 1910

Whitetop Mountain, Virginia, August 17, 1955. Under rocks at summit, 4-0-4. Under bark of log at elevation of 4700 feet, 0-0-1. Prof. W. A. Harman. Mountain Lake, A'irginia. Barn, Mountain Lake Hotel, June 20, 1955, 0-0-1. Moonshine Dell, June 20, 1955, 0-0-2. Castle Rock, July 8, 1955, 0-0-1. Prof. W. A. Harman. Mann's Bog, MLBS., August 4, 1955, 0-0-2. Doris Hatfield per Prof. W. A. Harman.

Segments, 133 (2 helicometameres in intestinal region), 136, 141, 144 (all juvenile), 141 (largest clitellate worm, 110 x 7 mm), 152. Pigmentation red, quite dense in juveniles and obvious ex- ternally clear to hind end, also externally recognizable all the way to the anal region in one clitellate worm but in others visi- ble just in the preclitellar portion of the body and even there sometimes indicated only Iw a faint pink. Pigment appears to be sparse in dorsum of ix-xii, except near mD, of the clitellate worm with most obvious red color. Clitellum whitish or light grayish and, even though in formalin, about as in previous alcoholic material.

Tubercula pubertatis as before, quite rudimentary but obvi- ously tripartite on two late juveniles. Variation in location of genital tumescences and clitellum is shown in the table.

The typhlosole ends in cv, but is smaller in last two or three segments (worm of 141 segments), cxiv (152 segments).

Seminal vesicles of a juvenile are large, filling coelomic cavi- ties except in ix, very soft and fragmenting on slight pressure even after a month in formalin (preservation very good). Testes discoidal. Iridescence lacking on male funnels.

Eemarks. The red coloration in clitellar epidermis previously mentioned (Gates 1955, p. 8) must have been slowly developed in the formalin. No trace of the coloration is as yet recognizable on these recently preserved worms.

GATES: AMERICAN LUJIBRICIDAE

19

Table 3

Variation in external characteristic's of E. ('(iroiiju'usis Itoui

Virginia

Genital tumescences,

including ah,

additional to those

Serial

on xxv-xxx, ioentcfl

Clitellum

number

in segments

on segments

Remarks

1

11

17 18 - 31

Juvenile. TP rud.

O

11

17 18 31

24-32/eq

3

11

17 21 24 31

24-31

■±

11

18 21 31

24 32

Sph.

5