CE Oe i en oe oy own) eters he Ae ee D.C nL ee ee a a ee Sea EL ea We ee SO ae Wo Bane Ne were gen 7 § Te ee YO a a ’ a) saa re, a ts ’ ore ba oN Ae eo lt “, es et ek hee er ee re ee ee tee ig ” oe Pe Pee en thedtn aot del RiGee! oS diash, bred Pk pee te Wage yee cbt q ’ . CRA LAS He tS Wael olay AN wane 327,00 LN A Teng Baw €. 10 AE gee AT Ne tat to ‘os me eatet Wa ee eee ae Bem bw ee es “0 * pow ge “4 BC EOL YE tre boy Re ea ® ee ee a Reeve ny to sary ty ioe oh Re meee be. 4s ¢ Wet 4" A vi ar a UO ee grit es Bel ety Sydhny es et Ch ce rr i ic, a ce Ree eh woe ee Toe Rela’ Ce Le 1 the Ne Vw REE We tea ee te pe PC ee Tee Obed ee beta hack ads Ads, Tota Mash » aA Ah ea. mile Mp ns fh ore ela oe Wan ag ee * eo Ee ee Rehm & Pe ee ek ld me bey Meee eee enw en ee ’ tae * LW Wy geen free OF Pm A Ce See ee A ee ee Pe 2 i ee J on ae ie tip chats Sokal Ae. Ce ‘ oe Or i ee. oe ie ea] em hen Pree a) tyes Aa ee a) ey wa ten « mbes sith yaa Re ay nas ah wwe were i eaee at oh ihe he oy a on}. ee oe ee er Cea i at i F afew fet . ae NOR ear ee or Haine wt seen Am aoe my wae abe SAAC A A ag VAL OP Got oD at Pee Se wn te Lele yeh Rg s o vie @ renth ve | Sree weet ke : A whe a een ee ete w% ae Ate ren Cat a et Hwee Ce cee, re eo ne eo rc Ted Oi mere ee eh ee ee Se Se 2 eee We 4 re ee ee Aras 4% er ee eo co ok | Sr oe re oy to ete CC ve ee er Cr ee a or 2 ees So oe wre Be eee [ee oe ae Ue Oe vel tele ee et LY se ae mt eee ee oe ee Pee eC On re te oo eet a restatyelh Bibs ne wees, oe ee er a he Se aT ee ed be oo ot ay ane DRL re en ee rn ey eo ve a ee A ere Bo ee ey ener ee ee eee toe me yi bey ” x bhp rs atom tex 6 by oe inte tg hey 4 Lt bee hho | * Heyy eee ee = ith are, ed Co ae Werke Vo goth Paidkae Te hy: 4M 48 ek a 44 eae Scene eh er Ce eC ee es Ce Oe eee ee ee wy et bei ke ee ge ren weer ee er ee ee 1 Oh Mate ee i ee LT wee Ee ak ew ath tat Maite ctoviech vm rape agtrrie oe feo dee Oops A ee BD ym eae | ae oe ek eee nlp a yA eh Sete ee heh Ct ee ee en we Ney er etal, Meatindh tek ticle Se OI ream Hi AEN Hg a ta baie Coen Cet er ee, oe en oo 1 Qa ew Wa ee CF he me be Taran ee ye at dp. * We bomen s SE AS mere ne OTD be om ete re ho 1 a he a a tet. hee Re ee ee eh eee rt Hoh a ere tee Cn Oe ee oo ho i oe Po eee ee oe eo te Veet ree mia ee eb ey Pe ee FO te mene re ee et hn eT Se be De a re Wie ee er ee ee ee »¥ swewy red ie ee ee ee es pe a Vile wae ee ee el cee ee cro hea Metre. APS A Oe th Pet te My es Se em Sr eh ee Cr eS. oe ok ot mee ; ta ty howe re ee ns A Ne LE ee Rm heme ee tt eee Fe ee a i a ee ern eo ihn ck eC od aoe eee 8 Fe ete ew Se A Bt eee AN he ere ee ce OE ee Se pe ee bse oe ee ee Pe es See PN Ai Nan on te nn hy eo ht ol rep tap 4 ny hee ae Ce ee en ee ta Cr, oe ee ee see te Sc ee ay rae one Wt Ae sme yn ee ee B Nate pene oe Man. SU er Weer Ee We eG ee Se se ae ea ee he ee Se Cre ee ee oe 3 ate : “ee hg ee eh ee wae " ~ ec ee pp a ae eC A oc i ear ne ren Nery i OO ee Ge en er tecacit re eet BN OT RAN Urey 8 he @ 405s ws aera eres 2 ty tat Bie meee > ee a oe oe ee er oY ‘ wea ‘ ” o ‘ ee en) v1 . Pp Swe tger eg ob fy tye 1 a8 swe oa ye Oe ee) ‘ ae .ee ay ' Fate} ee eS hoe wt re , ke Le we ee “y ' we ' ” Ph eee a ee ie ee aa ‘ Se ee yee eee ee ty na “ + RR be Oy he irem, 4 i> aby 4 . . ‘ ~t tvs ‘ Vane ve eH eh ‘ . ' tae te se re ‘ = * m4 oat ‘ reer ee ony ‘ : eye a Pa ae er Oe Pa ‘ , ‘ tbe ee ede Pe ee eam bens ait ony ‘ sees tee ow set Vw we oa) we tat . aks peered ' oan a Tae er re oe see Rarer votes Vis bets eet be a Se ey et Pee Wk ee ee on oe spe ts 4 8 Pes .» Pei ie eee SC ney ’ preee rary ay s4 ' er ee wont ek cd WR ee Hee YL ome eee e poe bet Pee De ee 2 ' wy hw Panu pied Ce ee | vy ton tonnes ee ee Mee ’ seed Cre ee, re ese ‘ ae ue peeve bea . wo . 4 Peep Vee ere aw & AYN tee , . ons ‘ 4 ere hese eae CC Ce ee Sg Whew it ve te pet 0] . ther eee 8 rem rr CC ce cre ee oc we . 1p ee Oe we ee ee A Coreen, P oa atari ees re ee ee ee ry ‘ v" Cr os ec ao ty tee ’ hoote ’ ar ‘ aoe th teed ee i ‘ > 44 an fee shat Thier -¢ oe bye Perera Or Wee 4 ta weal re +t pmeee NH . cee eee ' vt ete tee ny teat wanes eat ee we * ° ace an ee ‘ , roto ’ ( vet (grr daw +o bes os ‘tyes Pree ee re os en 4 ee See e+ . es . veiw OF . iis i pr gey Rat ty aes, hh ‘ vas , ‘ +2 es seer EW ot Desire. & eR ban tee ore ee ' ve ae rey . w Pie tee oy eet Paneer . ’ ' Ay wy ee ie ei « 4 Fue I de wT et « “ ah '- Tn) aes a ene ee , i 4 ‘ “1 . ‘ ' arate : - joe ees teh vew on ’ ar eee a ‘ ‘ » . ary . ae ’ ot ee ows Oe a ee ee ” Tis ° : ‘ 1 , ; a sues ray : ‘ . ot *¥ ' ' ‘ . © ype oa vbw s Biey wo Ce re . ‘ : ‘ , . Ving ‘ vis to . i qe ty _" Pars * . oe vee ‘ \ A an ray ‘ 7 ‘ 14 , ‘ - he's hae oa - . f Pe re ee ' ‘ peoates yore so ' ' ’ , \ hoe ‘ ’ . . 7 t ’ ‘ ’ hee ¥ ' P chia 1 ‘ , ° ‘ ‘ .* . : ey ’ ' . te n ' : ' ‘ Fs ri ’ pve ‘ 1 . 4 hayes * ' reew . , ‘ ' vy . ’ ‘ yo ” + ee Are ’ + ’ ' 4 ‘ a (RPL, ; ' : ‘ ‘ “ . . a . 4 e . . x an . ; > mi Be’ 4 . ' ' % . “ 7 . i 78 - ' + ar “ : tone + . . ’ 4 4 ‘ ‘ han * . is * .t- ) , te ‘ . : ; * . aes ‘ ' ' ” . ' ‘ bubs : ‘ } * ’ : : Tat he } we : yr ‘ ‘ b beg ‘ an - “at . . z ‘ 4 ‘ ‘ 7 = ; 3! t+ - ; 7 . ; - ; os e , » iat ’ ' ; ‘a 7 . “a ‘ i ; = ‘ ’ ' . : A P . i . 7 ' + Y . = ’ x i : : tat : ' ‘ i ide : FE : si 4 i ‘ sb : ' i ? ’ ' ' é rity i ' : ety it ‘ ‘ 4 ! : 1 7 . ue wo ew wl - . ‘ ‘ teu ‘ ou . ‘ r ’ ? ” F ‘ ar) ‘ Rhea ‘ ' : : 3 . 7 . + » +4 . fw a és - ’ ‘ ‘ 1” b hea ede sae ot 4 + ’ ' . ‘ : - ‘ y tye ~ ae - io i . v€ as , one oF ‘ s i ‘ ‘ . ‘ ss wood 4 i ‘ C ‘ “ . . ‘ oa “ ia 7 F ‘ ‘ Bee be , : . ’ ¢ “ we w ‘ : oe, i . ar rer 6 48 Ve . ae “6 ‘ . ? - +n # ~ - . t oa 1 ‘ * ie oe . . er ‘- ‘ ‘ ‘ ‘ ix ‘ er ow ‘ m ‘ eit 4 , can hee Col »beene ‘ ibe - ANNALS OF THE NEW YORK ACADEMY OF SCIENCES VOLUME XxXVI 1915 Editor ae ’ EDMUND OTIS HOVEY ES ya New York | Ee Published by the Academy 235 TOS 1915, 1916 THE NEW YORK ACADEMY OF SCIENCES (Lyceum oF Naturat History, 1817-1876) OFFICERS, 1915 President—GerorceE F. Kunz Vice-Presidents—CuaRLES P. BERKEY, RayMOND C. OSBURN, CHARLES BASKERVILLE, CLARK WISSLER Corresponding Secretary—Hernry E. Crampton, American Museum Recording Secretary—Epmunpv Otis Hovey, American Museum Treasurer—HENnyY J. CocHrRAN, 389 Fifth Avenue Librarian—Ratru W. Tower, American Museum Editor—Epmunp Otis Hovey, American Museum SECTION OF GEOLOGY AND MINERALOGY an Chatrman—Cuar.es P. BerKey, Columbia University Secretary—Dovucias W. JoHNnson, Columbia University SECTION OF BIOLOGY : Chairman—RayMonvd C. OsBurn, Connec ticut College, New London Secretary—Witi1am K. Gregory, American Museum | SECTION OF ASTRONOMY, PHYSICS AND CHEMISTRY , {4 Chairman—CuHar es BAsKERVILLE, College of the City of New York _Secretary—Ernest E. Surru, 50 Hast 41st Street i ard ¥ 7 Ae fy CONTENTS OF VOLUME XXVI Page PT EME Tice noises «6 cisleare sb eaves witcha ina PEt oe eh tars eters asia Sia ra eke ehkte i Abe RIE yal creer aaah 06 foo ue ahen sie ee anere Lica ca djlehareterale Seve le 8 vac vw o'e-eha ete © ii SNR MRT oes A Seas has Spe ah scant cs 3 oho c a oe wile lei atalenk clap aie ate a tobereieie: a ld bee be ede'e ili Mates of publication and editions of the brochures.............0....00.00- iii ee MN STEER VOLTS 289 che i 7os n> oo, wie eek fenahover Bier see.o, aneiei ah aval Oe enw, Sealed eves od) eles iv Geological Reconnoissance of Porto Rico. By CHARLES P. BERKEY. re He Spee DUNN Recerca ad cl.’ SPais are gatebentrratte, Sea ooal'ah st asta Si ahaP ONG Le cen area eevee wie 1 List of Greater Antillean Spiders, with Notes on Their Distribution. Oe ee SE GU ET Actor Vais. ilsie's, OR Meebel sya Where abla Ge eels s OEMS ese « 71 Volcanic Dust Veils and Climatic Variations. By HENrykK ARCcTOWSKI... 149 Statistical Study of Variation in Spirifer mucronatus. By CHARLES C. eGR rc ee ia ate ants ol aecuase’ os wha tee) « MUPRTe a rOhah Re Ray GAL: of & IO ws Ra AAW oats 175 Review of the Pleistocene of Europe, Asia and Northern Africa. By Pasa ATRETE ED OSBORN cairo sais wo o.0 che bo bi detea ward ne rine here kere 215 Present Status of the Problem of the Origin of the Tetrapoda. By Wi- Dee GHCoOny. CP late DV )iic. occas & od sccam clea eo sce clen Mee ore Siete ‘317 Biochemical Studies of Selenium. By Victor E. Levine. (Plate V)..... 385 igecords of Meetings, 1915. By Henry EX. CRAMPTON...........062c-c006 395 Peet tshe ot Mecemmber, 1915.0. 2c. eee oe ae bo eee cee ceee eee 463 a aR ne a Sc aS ee ke eh Shara Qiev dw win bie Gi bhi tebe algls G Suche ab ere iele's 475 DATES OF PUBLICATION AND EDITIONS OF THE BROCHURES Edition Pp. 1-70, 3 March, 1915 - 1450 copies Pp. 71-148, 29 May, 1915 1075 copies. Pp. 149-174, 21 June, 1915 1200 copies. Pp. 175-214, 30 June, 1915 1150 copies. Pp. 215-315, 30 July, 1915 1350 copies Pp. 317-383, 7 July, 1915 1250 copies Pp. 385-394, 18 October, 1915 ~1310 copies Pp. 395-486, 12 May, 1916 1000 copies iii LIST OF ILLUSTRATIONS Plates I.—Relief Map of Porto Rico. II.—Geological Reconnoissance Map of Porto Rico. III.—Generalized Geological Cross-sections of the Island of Porto Rico. 1V.—Right Pectoral Girdle and Limb of Sauripterus taylori Hall. V.—lIllustrations of Biological Reductions of Selenium Compounds. Text Figures Page Relief features characteristic of the interior ranges of Porto Rico........ 8 Partially destroyed dune sand deposits of the San Juan formation....... 11 Unconformity below the Arecibo formation................. «2. see 16 Typical shale occurrence... . 2. 2 -sie. soc ce « os liaise = hee eles 5 ee 18 Interbedded coamo limestone layers with massive tuffs................. 20 Diorite porphyry sills...... 2.6 c 006s ews See ees ce we be te Cee a ene ee aT. Photomicrograph of a typical thoroughly indurated dese ash: . s353a8 30 Photomicrograph of a typical weathered shale from Fajardo Playa...... SL Photomicrograph of a foraminiferal shale from the Bayamon-Comerio TOBE foc soeeicie a 08 os =m dele wuclon a wie tthe aie sinh wml Oe at a err So - Photomicrograph of the San Juan formation indurated dune sand....... 33 Shales and ash beds cut by a large irregular dike and sill............... 36 Overturned fold and crush zone in finely bedded shales on the Jayuya road near the summit of the range......:.......cce..:55 se 38 Crumpled shales as seen along the Jayuya road near the summit of the TARO 202. ce eo a ele dic eS we de )ataje yes wm delet a Soul Sm ei Sree 40 Cliff of the San Juan formation south of the lighthouse at Arecibo....... 44 Detail of the double structure in the San Juan formation of Arecibo..... 46 Playa plain and marginal ‘terrace.......2..0.c..-+-..-0.0+- 0.0 48 Structure beneath the marginal terraces..................0.20e==nee 49 Inner lowland near Bayamon...........-..20- ee 55 ErOn os. wee ce dc os ee wo tne Slee eet 2 eee Be ee Coal and Oil... sce ccs cen we awe eos ede Se 56 Tamerock .. 2.2. o000 ss 0eeensscdnceneeeeeee= =e 56 GUANO ...c ccc c cece ce wi cee eee es 6 ee i oe al eee tele 57 Road metal ....... 0.000 cce eens ccnns cee occe = ae ieee 57 Hot-Springs «oc i. See ss es ee ae ee eee PE 57. Historical statement <2 2:2 “Sects lose = eens eee eer 2 ow 58 Future problems. ....-...:..i..-cee2-ceee oe descdnss es oes ee 61 Base. map: is. 0is te. oe Se eee eerer 61 Geologic MAP .........0e ce ewe cee e tee ccc es eene eee wee ee ee District studies .. 063. cust cece ows ook eS bes banc orca 62 Reef-building organisms ......0.5.0. 0.00. c0s0ceu cee sm gee 63 Paleontology 220%... ss dane cde ae ee eee 2 oe Sacer 64 Tertiary subdivision. ... 2.5.0... 00.000 se ce ec oan os oe a eee 64 San Juan formation.» ... 2.502. 26 se < awn me auwer le ae ee 64 Subdivision of pre-Tertiary complex...........2...<.2.+s.00e—eeeene 64 Mineral reSOUFCES 2... oa ccc ec eek chee ae sem ood ee ane 2 eee 64 Petrography ... 2. seco eee we ls ed De ee Se ee eee 65 Physiography ......s000cteccest wen'es nce elec eae s cle Pres oF Thermal. waters... 0.3% Soest Sc ee ee eae ss a ee oe 65 Geologic history .........0. 2.0. se cecc ecu ce se e- © ccs ieee eee Collections ... 2.2.6... cso c Oeenwies cen eble Jecleecelenel SS 66 Tilustrations. ... 2... 5065. lec ad ce cen = demas be © aoe oe ee 67 Cross-sections . o.oo... jcc can cee cake ce cums enters one aeee an 67 Maps 2. 0. is nse Wubi ese otek ene ee je eee es oe 68 Acknowledgments <2 ss5552555208 deen I 68 Bibliography .... 220.60. cednn les Yee d oe a poe eee ee 69 INTRODUCTION The Island of Porto Rico has never had a thorough or detailed geo- logical study. There have been, however, a number of papers written that have described special features or general conditions in a very ac- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 3 ceptable manner. Most of the writings? of this kind are widely scattered in volumes of periodicals or proceedings of learned societies or pamphlets which in most cases are not readily obtained or consulted. From a perusal of these articles, one learns that Porto Rico belongs structurally and genetically to the mountain chain now represented by the isolated islands forming the principal West Indian group. Enough work has been done, especially by R. T. Hill? to outline roughly the geological history of the Island of Porto Rico and indicate on a map the distribution of some of the formations. An especially good general de- scription of physical features, also, is given by H. M. Wilson.* An intro- ductory general description, in large part along the same lines as these, will probably serve the present purpose. GENERAL DESCRIPTION. The Island of Porto Rico is situated in the Torrid Zone between lati- tude 17° 54” and 18° 30” north and longitude 65° 13” and 67° 15” west. It is the easternmost and southernmost of the Greater Antilles. It lies within the trade-wind belt, and the constancy of these winds gives the island a remarkably mild and uniform climate. There is an abundance of rainfall on the windward side, which in this case is the east end and the north side as far west as Camuy. The effect of the mountains across which these winds blow is to make the south side of the island and most of the western portion comparatively arid. Some districts are said to have no rainfall for a whole year at a stretch. The Atlantic Ocean lies to the north and east, the Caribbean Sea lies to the south, while Mona Channel on the west separates the Island of Porto Rico from Hayti. Brownson Deep, reaching the profound depth of twenty-four thousand feet below sea level, one of the deepest spots known, hes immediately to the north. Tanner Deep hes to the south, reaching a depth of fifteen thousand feet. Although the relief of the island above sea level is less than four thousand feet, this represents only the extreme top of a great mountain mass which rises above the submerged platform, from which its real height should be measured. The extreme relief differ- ence represented by the summit of El Yunque on the one hand and Brown- 2The writings referred to, together with others that have some description of physical conditions in Porto Rico, are listed at the end of this paper. Those of most usefulness in the present investigation are certain papers by R. T. Hill and H. M. Wilson, besides a very few others of less extended character. 3’R. T. Hitt: ‘Porto Rico.” National Geographic Magazine, volume 10, pages 93 to fae (1889). 4H. M. Wiutson: “Water Resources of Porto Rico.” Water Supply Paper No. 39, U. S. Geological Survey. 4. ANNALS NEW YORK ACADEMY OF SCIENCES son Deep cn the other is approximately twenty-eight thousand feet. Re- garded in this way, the Island of Porto Rico belongs to one of the higher relief features of the earth. It is in reality a badly eroded summit of a great mountain belonging to an east-west chain or range including the Greater Antilles. The gen- eral structural features of the islands are consistent with this east-west axial trend which is expressed in the topography of the central Cordillera extending from the west end near Rincon, where it starts abruptly from the water’s edge, to Fajardo, where it terminates in E] Yunque, the high- est point on the island. The mountain range, however, is not so simple as this statement would lead one to believe, for there are in reality two ranges or branches toward the east, one of which is known as the Sierra de Luquillo, culminating in El Yunque, and the other, which is best de- veloped in the divide between Cayey and Guayama, is called the Sierra de Cayey. The military road crosses this latter branch near Aibonito over a pass that reaches above 2,000 feet. ‘To the westward, the double character of the mountain ranges is not so pronounced, but there is a semblance of it in the spurs that reach the sea rather abruptly, one near Rincon and the other near Mayaguez. The exact elevations of the higher mountains have not been accurately determined, the values given on the older maps being evidently too great. .The revised approximate elevations are: for El Yunque, at the eastern end of the island, 3,750 feet; for El Guularte, which stands to.the west of the Arecibo road, 3,610 feet. The highest point in the Sierra de Cayey is about 3,000 feet. Many points are nearly as high as those given, and all of the roads that cross the island reach elevations on the divide that are in excess of 2,000 feet. Many of the roadways in the interior districts reach elevations over 2,225 feet. Although the island as a whole has a mountainous aspect, and although much of the interior is very rugged and picturesque, there is usually a comparatively gentle or smooth topography along the coast, and some of the marginal areas are almost perfectly flat. These are uniformly at the mouths of the larger rivers and represent river alluvium or delta-like deposits,—they are known in the island as playas. The aspect of the island as a whole is moderately rugged. Although the major portion of the rock makeup is igneous, and although there is considerable complexity of structure represented in all parts of the range, all of the surface forms are of erosional origin. The relief is that of early maturity in the interior and perhaps late maturity in portions of the coastal districts. Exceedingly steep slopes are the rule in all parts of the island where there is any considerable relief, and one of the most surprising things is the way the soil clings to these slopes. One often meee BERKEY, GEOLOGICAL RECONNOISSANCE OF SAN JUAN ee, RELIEF MAP OF PORTO RICO compiled by PORTO Role 3 drawn by O.A.LJUNGSTEDT Oty = E ABIV Via, Lee CaB RICO __ 20miles Porro Rico I.—Rruibr Mar or PLATE ) Agriculture, Department of S. TY ds 5 of the Forestry Division, l ) Reproduced by permission from Bulletin ‘ 6 ANNALS NEW YORK ACADEMY OF SCIENCES sees slopes of this kind with angles of thirty to forty degrees from the horizontal under cultivation. The islands of Culebra and rasa he to the east of Porto Rico and are said to be similar in structure and makeup, but neither of them was visited on this expedition. The streams of the island are numerous and surprisingly large for the size of the areas drained. According to Mr. Wilson’s description® of the water supply there are twelve large streams flowing north, four flowing west, seventeen flowing south and six flowing east. Besides these there are said to be 1,300 small streams, and on account of the heavy rainfall on portions of the island many of them are of larger size than such an area would usually afford. The main divide runs near the southerly margin of the island, so that about one-third of the drainage is tributary to the Caribbean Sea on the south, and about two-thirds to the Atlantic Ocean on the north. This unsymmetrical position of the dividing range is an abnormal feature, the cause of which is the subject of discussion in an- other portion of the report. Because of the prevailing trade winds, the rainfall is very unevenly distributed. The east end and the north side are comparatively humid and plentifully watered; in contrast, the west end and especially the south side are comparatively arid. In order to overcome partly the shortage of water, a large system of irrigation 1s now being developed on the south side of the island. At the east end, north- east of El Yunque, there is an annual rainfall of 123 inches. On the other hand, at Cabo Rojo, at the other extreme on the southwest side of the island, it is exceedingly dry, and in occasional years there is said to be not a single drop of rainfall. Other parts of the island vary between these extremes. The wettest months are September and November. In most districts, the underlying rock is compact enough to discourage much deep water circulation and the stream run-off is correspondingly responsive to the rainfall. In the northwest corner of the island, on the broad limestone belt extending from Aguadilla to Camuy, there is a pre- vailing tendency for the surface waters to sink into underground chan- nels, leaving the surface very much more poorly watered than even the somewhat scanty rainfall would lead one to expect. In some cases, streams developed on the more compact rocks of the interior districts completely lose themselves in underground channels upon entering the limestone belt, and in some cases do not again come to the surface for several miles. Elaborate caves and channel-like caverns are common and, in the north- ern belt of limestone country, there are thousands of such occurrences 5H. M. WILson: “Porto Rico; its Topography and Aspects.” Jour. Am. Geog. Soc., Vol. 32; p. 220; 1900: BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO Y still preserved that are now wholly abandoned by the waters that for- merly occupied and helped to form them. There are no inland lakes, but there are a few coastal lakes and they appear to be related to the develop- ment of alluvial plains or playas and recent elevation and subsidence changes. The climate is strictly tropical, but it is so tempered by reason of the constant trade wind breezes from the ocean and the elevated character of much of the ground that it is usually agreeable and mild. The lack of great changes of temperature and the prevailingly moist conditions on most of the island have direct influence on the character of the rock decay and disintegration and also on the quality of the soil produced as well as its behavior as a residuary product. Other matters of climatic conditions have little or no bearing on geological problems and may well be avoided. The average daily temperature is eighty degrees; it rarely goes above ninety degrees or below seventy. The maximum temperature is ninety- nine degrees. Extremes recorded for the year indicate a range of forty degrees. The area of Porto Rico is given as 3,670 square miles, which is about three-fourths the size of the State of Connecticut. It is roughly rectan- _ gular in outlime and in actual dimensions is about thirty-five miles in average width, and one hundred and five miles long from east to west. It is the fourth in size of the West India Islands and is one of the most productive and densely populated districts in America. Because of the greater interest recently taken in studying the natural ~resources and natural history of Porto Rico, it was judged to be a suitable time to make a more elaborate and detailed study of the island’s geo- logical framework and history. In accord with this view the New York Academy of Sciences organized an expedition which spent a part of the summer of 1914 on the island. The accompanying descriptions are based on the work accomplished by this expedition. New York ACADEMY OF SCIENCES EXPEDITION The geologists sent to make a preliminary study or reconnoissance of the Island of Porto Rico left New York on the 15th of August, 1914. Four weeks were spent in Porto Rico, the expedition returning to New York City on the 21st of September. The party consisted of Dr. Charles P. Berkey of Columbia University, New York, and Dr. Clarence N. Fenner of the Geophysical Laboratory, Washington. Arrangements were made with the bureau of transportation of the Insular government in San Juan for conveyances, so that as much ground as possible could be seen in the time available. More than 2,000 kilometers were covered by ANNALS NEW YORK ACADEMY OF SCIENCES CO the aid of this transportation service and observations were made in suffi- cient detail to judge the general character and structural relations of the formations crossed. In addition to this kind of travel on the roads, short trips were made on foot to examine features or outcrops of rock which appeared to deserve investigation, and an occasional more extended trip on horseback was taken to points in the interior. With these facilities for travel, it was possible for both members of the party to give undivided attention to geological observations. It was possible to stop and make Fic. 1.—Relief features characteristic of the interior ranges of Porto Rico Photograph taken from the Ponce-Penuelas road at K-10, looking northward across eroded formations of the older series to the main drainage divide. brief examinations along all the roads at hundreds of places, and, on several of the roads crossing the island, sufficiently elaborate data were secured to furnish a basis for geological cross-sections showing both relief and structural features. A complete circuit of the island was made and in addition it was crossed from north to south on three principal roads. This, together with numerous side trips into the interior, permitted ob- servations to be made on practically every formation of any considerable consequence in the island. No point in the whole area is situated more than seven miles from some road or other point of observation covered by the party, and, even in those cases, except in the extreme southwest : BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 9 corner of the island, observations were made on all sides or completely around the unexplored areas. Because of the extent of the reconnois- sance, it is the opinion of.the writer that all of the essential, large, funda- mental geological formational units have been found and their general structural relations have been determined. OBJECT OF THE EXPEDITION The purpose of this expedition was, first of all, to determine the nature and origin of the rock formations of Porto Rico, and to group them into series suitable for use in subsequent geological work. In the second place, it was the object of the party to determine as many of the larger struc- tural relations as could be seen and to unravel as much of the geologic history as such a hasty examination would warrant. In the third place, the physiographic features were studied for the purpose of determining their origin and relation to the formational structure and their bearing on the more recent history of the island. Lastly, it was appreciated that the island was complex enough to have many problems that could not be solved without very much more extended investigation, and it was the purpose of this expedition to point out the problems that should receive special study and that seemed to give promise of important results. Con- siderable attention has been given to the economic resources of the island by private individuals and considerable money has been spent on various enterprises connected with their development. These problems were also kept in mind, and wherever convenient, special observations were made on them. Although it is possible to make suggestions concerning these eco- nomic resources, they are for the most part matters that should receive very much more extended special study. A matter that concerns the wel- fare of the island more directly than any of these is the question of quality and variety and origin of the native soils. These of course are in large part geologic matters also, and although this reconnoissance is not sufficiently detailed to form the basis of a discussion of this matter, it is one of the lines of investigation connected with further work that will have direct value. INVESTIGATIONS AND DISCUSSION ROCK FORMATIONS The most fundamental thing to be determined at the outset of an in- vestigation of this kind is to discover and differentiate the different rock types and the structural units to which they belong. All of the prelimi- 10 ANNALS NEW YORK ACADEMY OF SCIENCES nary work of this expedition was devoted primarily to this question and collections were made for comparison throughout the island. In the beginning, observations were made chiefly along the coastal margins be- cause of the greater amount of ground that could be covered and because of the apparent simplicity of the outermost and younger or more recent formations. The chief formations with their representative rock variety will be discussed in order from the younger to the older series. In the first place, such a reconnoissance shows that there are two great series of formations separated by a marked unconformity. Both are somewhat complex, but in that respect the older series is very much more complex, both in range of composition and number of units involved and in variety of structural relation, than the younger one. In spite of this discrepancy, it is still the most convenient and useful division to make, and, because of the strikingly different characteristics of the two series and the great prominence of the structural break between them, there is no possible chance for mistaking this fundamental feature. The whole lot of formational units are therefore grouped under the follow: , two heads: 1) Younger Series. Including the Tertiary shales, reef limestones and recent deposits. 2) Older Series. Including a complex group of formational units,—tuffs, ashes, shales, conglomerates, limestones and a great variety of intrusives, all of which are probably of pre-Tertiary age. 5 There are several possible subdivisions of the younger series, but in this discussion only those exhibiting enough physical constancy and char- acter to be useful in field correlation are taken into account. These are especially (1) the San Juan Formation, a Pleistocene sand-dune deposit, and (2) the Arecibo Formation, a series of reef limestones and associated shales and marls. Besides these, there are more local developments that deserve special discussion, such as the San Sebastian shales, the Juana Diaz marls and sandy shales, and the Ponce chalk beds; but in a broad grouping these are all phases of the larger Arecibo Formation and it will take detailed paleontologic study to make the proper subdivisions. The older series has many formational members and their general rela- tions are reasonably well understood, but a systematic subdivision is not yet attempted. Correlation in this series is still more difficult than in the other, because of the great variation in character laterally and the influence of igneous activities that prevailed throughout its whole devel- opment. Some of the most characteristic of these types will be described. BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO {jt < Younger Series San Juan Formation.—On approaching the island by the usual route entering San Juan harbor, the first rock whose structural detail can be seen is that on which the city of San Juan itself is built. One can see that the formation is made up of strongly bedded material that has all of the structural characteristics of a cross-bedded sandstone and is resistant enough so that it forms, at this point, a promontory extending far beyond the supporting mainland and presenting a cliff face of at least 75 feet in height. A closer inspection supports all of these observations as to structure and adds the important observation that the granular material Fic. 2.—Partially destroyed dune sand deposits of the San Juan formation These deposits are a short distance west of Arecibo and are nearly covered with fresh j dune sands of the same material. is, in large part, calcareous in composition and organic in primary origin. The same type of material, with all of its characteristic structures, was seen at several other points along the north coast. Special studies were also made on sea cliffs of this formation in the vicinity ‘of Arecibo, where the exposures were so well developed that good photographs of the struc- ture could be taken. One of these is reproduced as an illustration of the characteristics of this formation. It shows exceedingly steep cross-bed- ding structure that measures up to thirty-three degrees dip, and there are also occasional structural lines that are almost horizontal. The rock is exceedingly porous, the grains are unusually uniform in size, and the 12 ANNALS NEW YORK ACADEMY OF SCIENCES binding material is calcareous, attaching one grain to another merely at the point of contact. The extension of ledges of this kind of rock far beyond the possible reach of swift-flowing streams, together with the fact that the distribution is limited to certain sections of the north coast, and, in addition, the evidence furnished by the internal structure of the rock itself, lead to the conclusion that the formation is essentially an old dune- sand deposit. Sand dunes are developed on the present coast line from very similar material, but none of the very recent dunes are solidified. On the other hand, the San Juan formation seems to have been developed before certain of the later elevations and subsidences that affected the island in its recent history, so that its material is fairly well cemented and its base extends below the present water level. Its outcrops also ex- tend to greater elevation above sea level than any of the modern dunes. It is judged, therefore, that this particular formation is the most recent of all in the island to act as a ledge former, and it is judged by its situa- tion and content to be of Pleistocene age. See additional description under heading “Structural Features.” This is the most unusual formation in the island. It is a type seldom seen or seldom recognized, and it is one of the smallest in Porto Rico, in spite of the fact that it makes such an important showing at San Juan harbor. Because of the prominence of the formation in the city of San Juan it is suggested by the writer that the name San Juan formation be used for it and that this name be confined to the Pleistocene beds representing solidified sand-dune deposits. Arecibo Formation.—Next below and older than the San Juan forma- tion is a great series of reef limestones and shell limestones preceded by shales that form a belt of considerable width along the north coast and a portion of the south coast of the island. In a large way this series forms a structural unit. Above it in all cases le the recent alluvial -deposits and the San Juan formation and below it le the older and more compl- cated igneous and sedimentary rocks. The break between these two rep- resents the chief unconformity in the whole geological column. The heaviest development of this formation is along the north coast between Tao Alto and Aguidilla. In this belt, the massive limestones of the Are- cibo formation attain the greatest thickness observed anywhere on the island, but no opportunity was found for determining the amount accu- rately. There is in sight, however, certainly as much as 500 or 600 feet in the bluffs along the Arecibo River. In this belt also, especially farther toward the west, in the vicinity of Lares and San Sebastian, there are un- derlying shales of considerable thickness which in places carry lignitie BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 413 material and which have been the object of some exploratory work for coal. It is evident, from observations made, that the shale beds of the Arecibo formation vary greatly and in some places are entirely wanting. At the Arecibo River, for example, where the beds of the formation can be seen well exposed in the river bluffs, there is no shale development at all. ‘The limestones he abruptly on the eroded and somewhat weathered surface of the older formations which at this point are represented by coarse and obscurely bedded or even massive volcanic tuffs. At the best place seen, which was on the east side of the river about opposite K-66 on the Arecibo road, there were a few feet of transition material between the limestone beds proper and the unmodified tuff. It had not the structural appearance and makeup of the shale as seen at other points, however, and this part of the formation is regarded as entirely absent on the Are- _¢cibo River. Shaly beds, however, are seen again on the south side of the island and their best development is in the vicinity of Juana Diaz, where some of the beds are distinctly sandy and rather fossiliferous and carry petrified wood. Lignite is also reported from this vicinity, but no ma- terial of that kind was found by the writer. It is more than likely. that the shale beds on opposite sides of the island do not correspond in horizon at all, but that structurally they are both basal beds. The most striking development of the shale beds and overlying marls and softer layers of thin-bedded character, instead of the massive reef structure, is on the Jacaguas River south of Juana Diaz. The dips also of the formation in this particular locality are much greater than those observed at any other point. For considerable distances an average dip of 30 to 36 degrees was estimated and the total thickness represented, based upon the width of the belt, must be at least 3,000 feet. At no other point on the south side, however, was there an opportunity to see whether the beds of this character are constant or of large lateral extent. As one goes eastward, a comparatively short distance, they are almost en- tirely lacking. On the Descalabrado River, which is only ten kilometers to the east, the underlying older series of tuffs and intruded shales and limestones were followed to a distance of two miles south of the military road, whereas at Juana Diaz the basal shales of the Arecibo formation begin a half mile above the military road. It appears, therefore, that the formational margin is swinging rapidly southward, and it is judged, from other observations made, that there is almost nothing of it repre- sented at a distance of twenty kilometers to the east, or, in other words, that the formation does not extend farther east than the vicinity of Sa- linas. At one other point on the south margin of the island, there is an unusually good opportunity to follow the successions of formations, and 14 ANNALS NEW YORK ACADEMY OF SCIENCES that is in the vicinity of Guanica. Limestones belonging to the older series occur immediately south of Yauco and are very strongly developed there. The hills in which these beds outcrop extend southward almost continuously to Guanica, but at that point observations showed that the formation had changed and is actually part of the reef limestone of the Arecibo formation, although it is possible that the large fault observed west of Ponce may pass through this area and obscure the other structural relations. It looks, from the rapid survey, as though it would be favor- able for some of these additional studies of the character of the lower beds of the Arecibo formation. Between Guanica and Juana Diaz, wher- ever the inner margin of the Arecibo formation was seen, it was bounded by a fault which brings the upper beds abruptly against the older tufts and shales of the pre-Tertiary. The formation furnishes an abundance of fossils. The lower portion | or the lower beds on the south side of the island, as seen at Juana Diaz, seem to be the most promising for a determination of the age of the beds of the. formation. Higher beds, forming a chalky white limestone to the west of Ponce, are also very fossiliferous, but in this area they are sepa- rated from the older rock series by a fault, so that it is quite impossible to tell how far above the base of the formation these beds may lie. It is judged that the portion of the formation seen at Guanica is a still higher horizon, but the exact age values have not been worked out. The total thickness of the whole Tertiary series on the south side of the island is very great. It was estimated that the shales and marls and limestones in the vicinity of Juana Diaz must certainly amount to three or four thousand feet. For a long distance along the Jacaguas River south of Juana Diaz, the beds stand with a dip of approximately thirty-five de- erees toward the south and throughout the greater portion the character is notably different from the beds occurring farther to the west which are judged to overly them. | To the east of San Juan, along the north coast, there is much less prominence of the Arecibo formation and after passing Rio Piedras it in no place crosses the main road. ‘There are occasional hills somewhat similar to those characteristic of the landscape of Bayamon and vicinity, but they do not reach to so great a height and are separated by very much larger stretches of low ground. The strongest development of this for- mation seen to the east of San Juan is that along the Grande de Loiza River between the railway and the coast. In going still farther east to the vicinity of Luquillo, the inner margin of the formation passes out to sea and the older formation reaches the shore. From this point around the whole eastern end of the island no more outcrops of the Arecibo for- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO. 15 mation were seen, and this is true also of the southerly side coming from the east end through Naguabo and Yabucoa and Guayama, and still far- ther west to some undetermined point between Guayama and Santa Isa- bel. There may he, however, occurrences of this formation underlying the alluvial material along the coast at points considerably nearer Guay- ama than the outcrop map indicates. The formation as a whole is essentially a structural unit. Although it is quite easy to distinguish the underlying shale member, especially well developed between San Sebastian and Lares, and although there are other structural changes, there is nowhere any appearance of unconform- ity or marked break in the succession. Besides, the shale member is not everywhere developed and, as a matter of fact, is seldom seen in tracing the formational boundaries. In some places, it is definitely shown to be absent, so that it seems unwarranted to represent this member as an im- portant part of the formation, so far as areal distribution is concerned. Judging from field observations already made, the shale and mar] beds are more extensively developed on the south side than elsewhere. But this is based on observations in one particular area and the member can- not be traced very far in either direction because of other difficulties. The beds lying above the shales and representing the part referred to as a more massive limestone portion are probably susceptible to considerable differentiation on the basis of fossil content, and it is entirely possible that a rather complete range of Tertiary horizons may be determined after complete paleontologic study. In the field, however, and on the basis of structural factors, there is no apparent ground for subdivision. In this discussion, therefore, the whole series of beds, from the uncon- formity at the base to the alluvium and San Juan formation overlying it, is referred to as a single formation and, on account of its extensive devel- opment in the region about Arecibo, it is suggested that a suitable name would be the Arecibo formation. _ Some parts of the formation show the peculiarities and content of a coral reef, and these portions have the most irregular and most massive structures. Other parts show bedding structure more or less perfectly de- veloped, and throughout the whole formation here and there, at irregular intervals, and usually of only very limited extent, there are more shaly facies. It is the opinion of the writer that this irregular distribution of shaly beds is responsible for one of the peculiar topographic features de- veloped in the belt underlain by the Arecibo formation. This is the occurrence of almost perfectly flat soil-covered areas of no very great lateral extent at different levels, above which the numerous small knobs or hillocks of limestone rise, giving the peculiar haystack-like topography. 1G. ° ANNALS NEW YORK ACADEMY OF SCIENCES This type of topography is represented by the small level tract surrounded or dotted over with small hills, called “pepino hills” locally, standing like haystacks above the plain at many different elevations above the sea. This leads to the belef that the fundamental control in its development is the presence of a shaly bed mm the series, which forms at each poimt the basis of the local plain. The hillocks standing above it or surrounding it represent remnants of the more massive and probably more porous and more easily destroyed limestone which has been attacked and largely removed by weathering, and especially by solution, down to the more " £ ites: SY ta es -. Fic. 3.—Unconformity telow the Arecibo fermatian This view shows the older tilted and eroded tuffs and shales below, as seen on the Arecibo River. The contact is immediately beneath the horizontally bedded limestones at about the center of the view where the chief weathering is noted. resistant shaly soil-forming member. If this is the principal cause of the peculiar topographic form just described, it is quite easy to see that the distribution of such features should be expected to be rather irregular both in Jateral extent and in actual elevation above sea level or in rela- tion to the different horizons in the formation itself. In regarding this as the principal factor, there is no tendency to overlook the fact that the island has stood at different elevations with respect to sea level in former times, and that a corresponding difference In ground water levels would be felt throughout the border region. But there is no evidence whatever BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 17% ‘that the numerous different levels represented by many of the small plains referred to above were necessarily connected with any of these subsidence changes. Additional comment on the hillock topography, so strongly developed in some parts of the area, is made in connection with discussion of drainage in another part of this paper. The Arecibo formation is of Tertiary age. So far as identifications of the fossils have gone, they appear to confirm the opinion that the larger part of the formation belongs to the Oligocene epoch. These determina- tions were based largely on collections made in the heavy limestone beds and reefs in the vicinity of the Quelbradillas River. The shale beds lying at the base of the series, and exposed farther to the south in the vicinity of Lares, are certainly somewhat older and probably belong to the Eocene epoch. There are higher beds developed rather irregularly that doubtless represent still later time, referred by Hill to the Miocene epoch, but these determinations must be left to future detailed study of the formation as a whole. It is considered eminently fitting to refer to some of the chief varia- tions which have especially strong development in certain localities, by special locality designations, such as,—San Sebastian shales, Ponce chalky limestones and marls, Juana Diaz shales and marls, Guanica coral reefs, Quebradillas reef Hmestones, etc. The correlating of all these and other local representatives of the Are- cibo formation is a work that can be done only by extensive and detailed stratigraphic study and paleontologic comparison. This is one of the larger pure-science problems awaiting future investigation. Older Series Below the Arecibo formation and forming the surface in the interior, beyond the Arecibo margin, the island is made up of an exceedingly com- plex series of many different kinds of rocks. They include chiefly varie- ties of igneous rocks, both extrusive and intrusive, both fragmental and massive, ranging from small stringers or dikes or flows to large boss-like masses that cover many square miles in area. In addition to the igneous rocks of these types, there are numerous shale beds and conglomerates of rather massive habit aggregating a very great thickness, and with them are associated limestones and foraminiferal beds of considerable variety. A study of the rocks of this series for the purpose of determining their character and origin indicates that practically everything in the older series except the limy portions of the shales or the limestones proper are more or less directly of igneous origin. The coarser materials and those least affected by any secondary processes are the tuffs which are of direct 18 ANNALS NEW YORK ACADEMY OF. SCIENCES voleanic origin and are exceedingly abundant and extensive. They are found at intervals in all parts of the series, and it is impossible to say that they are either more or less abundant in those portions which appear to be older, rather than in those which appear to be younger or higher in the series. ‘The closest associate of these materials is the bedded tuff, made up of volcanic fragments which have been somewhat assorted by surface agencies so that they exhibit some sedimentary structural charac- teristics. These are also exceedingly abundant and widely distributed and they pass by insensible gradations of finer and finer materials into those Fic. 4.—Typical shale occurrence This is seen along the road between Ponce and Penuelas at K-10. The beds at this point lie in a less disturbed attitude than is usual in members of the older series. that are recognized as true ash beds. Most of these have become so thor- oughly cemented, or so much modified by secondary attack, that they now present a perfectly sound and compact appearance. In thin section, however, it is easy to see that the material is wholly voleanie and that the bedding is the only secondary modification except that having to do with the binding, induration or alteration of the rock. The ash beds are probably close relatives of the so-called shales. Shales——Rocks of this type are developed characteristically at Fa- jardo, at Mayaguez near Baranquitas and at numerous other points, espe- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO {19 cially along the divides toward the west. At the two points first men- tioned, in particular, they are light yellowish or reddish in color, rather porous in structure, strongly bedded and have a prominent blocky frac- ture habit. The exact character of these rocks is a question under study at the present time, but enough has been done to show that, in their present condition, they have been oxidized to the yellow or red color, and have been leached so that they have a porous structure and hght weight due to the removal of at least a part of some constituent that is more readily soluble than the rest of the rock. Microscopic comparison with beds of the same structural relation, but of very dark color and very dense habit, leads to the opinion that the two types are not essentially different in origin, but that the lighter colored and lighter weight shales, such as are found at Fajardo and at Mayaguez, are simply the weathered equivalents of darker ones. It appears from this comparison that the shales are normally highly calcareous and that the lime content is sup- plied by the presence of a very large amount of organic matter in the form of foraminifera. In some cases this organic matter makes up fully one half of the rock and in all cases weathering produces a very porous effect that should be expected to be identical with the red and yellow shales occurring typically at Fajardo and Mayaguez. The siliceous con- tent of all of the shales examined proves to be exceedingly fine and wholly lacking in granular or quartzose material such as characterizes most sedi- mentary shales. It is the judgment of the writer that this material in the shales of Porto Rico, instead of being the ordinary disintegration products derived from the weathering of ordinary land masses, is in reality largely ashy material of volcanic origin. With this conception of them, it would appear that even the limy shales are therefore close rela- tives of the ash beds, and it is entirely possible that they do not represent any great difference in history, but rather somewhat different surround- ings during accumulation. . Limestones.—Besides the shales, there are massive limestone beds of several different types. In most cases the occurrences are separated by structural complexities that make it uncertain about field correlation, but undoubtedly later field study will connect some of these and additional paleontologic study will arrange their succession. The most prominent occurrences seen are de- ‘seribed below. Ooamo Tuff-Limestone-—The limestone with the closest genetic re- semblance to the types already described is represented in a broad belt passing from south to northwest across the upper end of Coamo Reservoir ~ near Coamo Springs, and which can be traced in prominent development 20 ANNALS NEW YORK ACADEMY OF SCIENCES westward across the Descalabrado River to the Jacaguas Reservoir. Sim- ilar limestones are found at other points on the south side of the island and are judged to belong to the same member of the older series. Be- cause of the strong development in the vicinity of Coamo Springs, and because of the fact that it represents a type so striking as to be recog- nizable as a field unit, the limestone has been called by us in the field the Coamo Tuff-Limestone. It is developed in the vicinity of the Coamo Reservoir to a thickness of several hundred feet and its most character- istic appearance is the brownish mottled color effect produced by the Fic. 5.—IJnterbedded coamo limestone layers with massive tuffs This formation is seen at the military road crossing of the Descalabrado River. presence of fragments of tuff and accumulations of ash. In some beds this material is so abundant as to make up almost the whole rock and it becomes an interbedded tuff layer. Occasionally the limestone beds are very pure and almost entirely free from volcanic materials, and there are also numerous beds of real volcanic tuff, but typically there is an inter- mixture of tuff materials with the lime in great enough abundance to give a brownish spotted or mottled effect. An equally characteristic feature of the rock is its concretionary or nodular appearance due to algous growths to which the lime accumulation is chiefly due. The finest development to be seen anywhere in the island is on the Descalabrado BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 2{ River at the point where the military road crosses it. A photograph of the interbedded relation of tuff and limestone at this point accompanies this discussion. This is probably, in part at least, the rock referred to by Hill in some of his discussions as “mountain limestone.” It seems to us that the several occurrences of limestone beds which have clearly differ- ent relations in the series, and the probability of being able to place them in different horizons, based on this content, makes it desirable to use more characteristic local designations for them. Such usage cannot be confusing even if it should be proven, as it may very well be, that some of them are identical. In this discussion, therefore, wherever possible, the chief occurrences that are not clearly identical with formations already described are characterized by adding the name of the locality -where there is especially good development of the formation. BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 25 Tuffs—The most abundant of all of the rock types is represented by a great variety of volcanic tuffs. These accumulations seem to be the funda- mental basis of the whole island. Whatever has been formed in the way of shales, sandy beds or conglomerates or any other of the ordinary sedi- mentary types seem to be directly derived from the same material that makes up the tuffs and the rocks directly associated with them, related in an interbedded succession. In most cases, the tuffs are essentially massive in their structural habit and are made up of a complex accumulation of large and small fragments of volcanic materials which prove to be chiefly andesitic and closely related porphyritic types. There is great variety in texture and minor structure and present condition, but on the whole there is enough uniformity of actual composition to justify classing them all as andesitic tufts. , In very many places there is obscure bedding structure indicating a tendency to assort and work over this material at the time of its deposi- tion. This is especially noticeable in the finer materials and some of these beds are made up essentially of ash. These ash beds resemble the type referred to as shales so closely, in some places, that it is impossible to distinguish between them in their field appearance. I judge also that there is practically a gradation from one rock to the other, the ash beds showing transitions to shales, especially where they have developed under conditions encouraging much weathering and working over of the frag- mental materials and promoting the growth of organisms in sufficient amount to make the accumulating beds somewhat calcareous in compo- sition. Tuffs and ashes are well known to be especially lable to attack by alteration and to the ordinary changes that modify rocks. It so happens, therefore, that many of these representatives are completely modified and have become so dense that they exhibit none of their ashy or fragmental structure without microscopic examination. In this condition they are usually also exceedingly hard and as resistent to destruction as the hard- est crystalline rock. The largest development of massive almost struc- -tureless tuffs which were seen occupy the Sierra de Cayey between Guay- ama and Cayey and also the range along the military road toward Aibonito; but there are extensive occurrences in many other sections. Some of the most prominently developed bedded tuffs and ash beds were seen on the north flank of El Yunque along the Sabana River and on the north side of the divide below Comerio, and also along the Ponce-Arecibo Road both near the summit of the range and farther to the north midway between Arecibo and Utuado. From what was seen of this type of rock, it was not possible to form a 26 ANNALS NEW YORK ACADEMY OF SCIENCES definite conclusion concerning the age represented except by their rela- tion to certain interbedded shales and limestones. It appears that the underlying older portions of the series of tuffs and ash beds have com- paratively little of such interbedded calcareous material and have every- where, been modified or altered or metamorphosed to a greater degree than beds that lie higher in the series. But beyond this there is little to judge of the actual age. As one goes higher in the series, however, there are occasional prominent limestone members with which tuffs are intimately associated or interbedded, and it may be possible, by reason of these rela- tions, to form a more accurate estimate of the geologic age of this later portion of the series. Volcanic Flows.—In addition to the sedimentary beds of various sorts and the related tuffs, there are at occasional places evidences of volcanic lava flows. These were seen at several places on the road between Baya- mon and Barranquitas. They are amygdaloidal in present habit and represent vesicular basalt and andesites. On the whole, evidences of lava flows on a large scale are wanting. This kind of product seems to have been very much more rare than the fragmental type. A more prominent thing as a structural feature is the occurrence of very numerous intrusive bodies. Intrusives.—The intrusive masses in Porto Rico occur in all parts of the island and in all of the formations except the Arecibo and the over- lying alluvial deposits. No such evidence was seen in any part of the Tertiary of younger series; but the complex series of rocks representing the pre-Tertiary, here referred to as the older series, are cut in all sorts of ways by both large and small intrusive masses. The smaller intrusives are chiefly andesite porphyry in composition and have everywhere pene- trated the shales and ash beds. The commonest occurrence is in the form of small sills or sheets conformable to the bedding structure and varymg in thickness from only a few inches to many feet. These sills are so per- fect in form, have so little disturbed or modified the adjacent beds, and are so similar in general composition and appearance, after weathering, to the associated sedimentary beds, that it is quite impossible to deter- mine in all cases how much intrusive and how much original sedimentary rock is involved. The only thing noticed as a rule is the uniformity of petrographic structure that seems to be characteristic of the intrusive as compared with the associated beds. The simplest occurrences of sills of this kind, which at the same time show their igneous intrusive character, were seen near Fajardo, near Rio Piedras and in the vicinity of Comerio. But occurrences of the same kind are exceedingly numerous in nearly every district and in total amount form a very great additional thickness BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 27 to the bedded rocks, shales, ash beds, ete., with which they are associated. In some cases, these invading magmas have incorporated great quantities of fragmental matter, giving the rock in its present condition a very strikingly fragmental appearance. This habit associated with its per- fectly apparent intrusive relation makes a very unusual combination in the field. In many places there are included blocks of immediately adja- cent rock such as one sees in the occurrence at the quarry at Fajardo Playa, but, in extreme cases, the mass is chiefly fragmental in its make- up and one could not readily interpret its history without complete field Fic. 6.—Diorite porphyry sills These sills are intruded between layers of calcareous shales and ash beds on the road near Comerio. The streaked or banded layers are shales, the massive portions, seen best at the left side of the print, are sills. A transgressive relation can be seen between the two layers at the extreme left. determination of its relations. Such occurrences may be seen in the vicinity of Guayama on the road about a mile east of that place, and also a short distance south of Rio Piedras. Few of these intrusive masses show anything but a rather uniform medium grain texture and larger ones have the average appearance of a rather fine or medium grain diorite. The granular appearance, however, is probably deceptive, due to the way the rock disintegrates, for thin sec- tions made from many typical intrusive representatives are nearly all plainly porphyritic in texture. 28 ANNALS NEW YORK ACADEMY OF SCIENCES In addition to these intrusive masses, which are of small size or at least of not very great areal dimensions, there are in a few districts large boss- like occurrences of massive coarse-grained igneous rock. The boundaries of none of these have been traced out. but it is certain from the distribu- tion now known that in each case the area occupied is several square miles in extent. The most prominent one of this type of intrusive mass is that seen in the southeast portion of the island, including the district about Huamacao and Yabucoa and Las Piedras and Juncos. . Whether or not this is all one mass belonging to a single intrusion has not been deter- mined. The variety of composition seen in the different samples taken at different points is consistent with the presence of more than one intru- sive unit; but it is also possible and quite as likely that the variety ob- served is wholly due to magmatic differentiation. The southerly portion of this mass, especially that near Yabucoa, is represented by a very coarse, very guartzose and almost pegmatitic granite. Farther to the north, in the vicinity of Las Piedras and Juncos, the rock has the appearance of a syenite. Although a part of the rock does show the composition of a true syenite, by far the greater number of specimens collected on this expedi- tion show the presence of quartz in sufficient amount to make the rock a granite. It would appear, therefore, that this occurrence in the south- east portion of the island is essentially a granite mass and that it is of unusually large size, reaching practically from the coast at Maunabo to Caguas. The distance across this mass is, therefore, not less than about 12 miles north and south. In all probability it is not of quite so great an extent east and west, but these boundaries are unknown. One other large intrusive mass was observed in the west central part of the island, in the vicinity of Jayuya and Utuado. Im general appearance and texture this rock, in the average outcrop, does not differ much from that seen at the east end of the island which is commonly referred to as syenite. In this occurrence, however, such specimens as have been exam- ined with the aid of the microscope, show the presence of quartz in most. cases In sufficient amount to make the rock of granite composition.- In this case, as in that referred above, there are considerable differences of composition shown by the rocks which seem to be a part of the same mass. Specimens found, for example, near Adjuntas have the compositional characteristics of diorite, whereas a specimen taken near the margin of the boss on its northerly side, near K-53 on the road toward Arecibo, is a syenite. At certain other points near-Utuado, the rock is a granite porphyry. The best idea of the variety of composition and textural quality repre- sented by all kinds of intrusives in the island can be gathered from an BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 99 ww examination of the water-washed pebbles in the stream beds that have come down the steeper mountain sides. An examination of such material shows an extremely large varietal range, and, although by all means the greater number are some variety of diorite or andesite porphyry, there are occasional more basaltic and more acid types represented. Summary.—From this description of the variety of rock types repre- sented by the leading field units, it may be readily seen that a subdivision into members of mapable constancy is no easy matter. It is the opinion of the writer that, for the present work and for in- vestigations immediately to follow, local designations or names will be of most direct usefulness, and that a correlation should be expected to be the final outcome of a series of such studies. For the present, therefore, it is judged best to use the terms Fajardo shales, Mayaguez shales, Jayuya road shales, Barranquitas shales, etc., as suitable names in these respective districts for the shale member of the older series, without any intention of suggesting their equivalence. In spite of the physical similarity in these cases, it is not at all likely that they belong to the same horizon. Similar argument will hold for most of the other members,—the ash beds, the tuffs, the limestones and the conglomerates,—and it is recom- mended in these cases, also, that locality designations be used in the field investigations. Those described in this report are not necessarily all that deserve special designation, but the same rule may be applied to addi- tional occurrences without in any way obscuring the ultimate solutions of the problem of correlation. On account, therefore, of the complexity of the structure and the limited amount of work vet done in connecting the separated occurrences into continuous field units, it is advisable to retain such terms as Coqui limestone, Coamo limestone, Corozal lime- stone, La Muda limestone, Trujillo Alto hmestone, Cayey tuffs, etc., for the earlier descriptions and special studies. PETROGRAPHIC RANGE There is an exceedingly great variety of certain classes of rocks in the Island of Porto Rico. Those most prominently developed and showing the greatest variation in minor character, structure and relationship are the volcanics, especially the volcanic fragmentals. All sorts of tuffs, cin- der beds, ashes, mud flows and bombs are represented in great quantity, in very wide distribution and in all stages of alteration and induration. Observations made on thousands of occurrences of this character of materials leads to the conclusion that most of it is essentially of andesitic composition. Although there is an occasional fragment of either more basic or more acid composition, the predominant types are always of 30 ANNALS NEW YORK ACADEMY OF SCIENCES andesitic makeup. ‘The present condition of these rocks, representing as they do nearly all stages between fresh material and either a thoroughly weathered or considerably metamorphosed condition, is a more interest- ing study than their primary composition. Some of the most dense and resistant rocks in the whole island are these older metamorphosed tuffs and ashes, : Next in point of abundance is the group of crystalline igneous rocks. In this case there is somewhat greater prominence of varieties represent- Fic. 7.—Photomicrograph of a typical thoroughly indurated andesitic ash, magnification 28 diameters A rock of this type appears in the field as a dark-greenish hard resistant obscurely bedded layer, usually closely associated with more massive tuffs on the one hand or more strongly bedded shales on the other. The clear grains are mineral fragments: the more complex grains are fragments of lava, cinders, glass, etc., all thoroughly bound into a complex aggregate. ing the acid and basic ends of the classification scheme, but here also the rocks of the andesite-diorite family are by all means the most numerous and most widely distributed. The greater number of occurrences are represented by members of this family belonging to intrusives that would be classed normally as andesite porphyries; porphyrites of various kinds and diorite porphyries. The minor variations represented by these rocks 3 BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 31 would probably include all of the habits known to this family. In addi- tion to the porphyries of this family composition, there are less common occurrences of felsite, quartz porphyry, granite porphyry and_ basalt porphyry. In surface flows there is, besides the andesites, an occasional amygdaloidal basalt, but so far as observed there was no rhyolite or other very acid surface type. Among the massive larger intrusives, the com- monest and most abundant type is a granite porphyry or granite varying in some parts to the composition of syenite. A massive rock of the na- Fic. 8.—Photomicrograph of a typical weathered shale from Fajardo Playa, magnification 28 diameters The dark field is made up of an extremely fine aggregate of earthy materials. The white circular and irregular spots are entirely empty and constitute the porosity of the rock. The circular forms of these yoids suggest that they represent former calcareous content in the form of foraminifera, now completely removed by weathering. ture of a diorite is also represented, as is a very coarse rock of the nature of a giant granite. How these are related, how many intermediate varieties there may be and whether this variation represents magmatic differentiation within a single mass or instead different units of intru- sion, has not yet been determined in enough detail to make a positive statement. But in at least two cases where these large masses were seen, ANNALS NEW YORK ACADEMY OF SCIENCES Qo Oo the hasty examination given to them leads to the belief that differentia- tion effects can be traced. The closest relatives of the igneous rocks are the sediments, and be- cause of the fact that the material constituting these sediments has been furnished by the volcanic fragmenial supply in large part, their character and makeup is in many cases not strikingly different from the ashes and finer tuffs. They do, however, represent an additional assorting, an ad- ditional weathering and an additional opportunity for intermixture of Fic. 9.—Photomicrograph of a foraminiferal shale from the Bayamon-Comerio road, magnification 28 diamciers The dark areas are chiefly earthy aggregates of very fine texture: the whitish areas are calcareous spots which in many cases still preserve the forms of foraminifera. It is the removal of such materials from the shales that is believed to account for their porosity as seen in weathered outcrops. materials from different sources and of organic material developing at the same time. These conditions give a great range of composition and mineral makeup to the shales and sandstones and they merge by imper- ceptible gradations from simple tufaceous or arkosic sediments to eal- careous rocks or even to fairly pure limestones. The common source of the caleareous element in these rocks is from an intermixture of fora- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 33 miniferal matters representing organic growths accompanying the accu- mulation of the deposit. | The limestone members on the one side representing almost pure or- ganic accumulations, and the conglomerates on the other representing the simpler destructive volcanic fragmental matters, give-the range between which an exceedingly great variety of sediments are represented. The shales of the younger series, represented by the Lares and the Juana Diaz shales, are more strictly detrital and of true erosional and Fic. 10.—Photomicrograph of the San Juan formation indurated dune sand, magnification 28 diameters The clearest grains are simple mineral fragments; the dark ones and the grayish ones with internal structure are fragments of calcareous organic growths. The grayish matrix is a secondary binding material of calcitic composition, in this case practically filling the interstitial spaces. destructive origin, and in places they contain lignitic material which suggests different physical conditions. This shows, however, in their upper layers an increasing amount of organic content also, and finally are succeeded by limestones of wholly organic makeup. The failure of volcanic activity during and subsequent to that time gave no opportunity for the amount of intermixture that is seen in the shales of the older series, so that as a result the younger series of rocks is petrographically 34 ANNALS NEW YORK ACADEMY OF SCIENCES more simple and less modified. The organic content is of greater variety, however, and because of the reef-building tendency the primary structure is more varied than is seen in the limestones and shales of the pre-Ter- tiary representatives. The most striking petrographic type is, probably, the solidified dune sand making up the San Juan formation. Its uniformity of grain, strong cross-bedded character, porous habit, together with its rather sur- prising stability, make it an object of some considerable interest. Sev- — eral of these classes of rocks, therefore, represent petrographic series of unusual range and variety, and because of their perfection of develop- ment would seem to. warrant detailed study. There are no foliated metamorphic rocks so far as yet seen in Porto Rico. One specimen of such rock, a mica schist, was shown to the writer as having come from the Portuguese river not far from Ponce, but a hur- ried reconnoissance in the vicinity failed to uncover anything even re- sembling it. The specimen probably does not belong to the rocks of Porto Rico. No evidence whatever has been seen of conditions that would be expected where such rocks occur. | The most profoundly modified rock observed is a massive serpentine. Such material was seen at two places by this party, one near Yauco and another on the road to Comerio, and the same type is reported by Pro- fessor Crampton on a much larger scale near Maricao. But they are al! simple petrographic cases after all, being ordinary intrusive units of heavier ferro-magnesian content than the average which have been heay- ily altered, especially by hydration processes, to the present condition. Genetically and historically, the serpentines are not materially different from the other intrusive bodies. Depth of Decay Alteration has affected the rocks at most points to considerable depth, but in spite of this there are plenty of outcrops, and along the roads there are many cuts exposing fresh rock. The stream beds are strewn with fresh pebbles and boulders. Although decay obscures the character of the rock in most of the outcrops, the structure is usually fairly well pre- served, and in most cases enough can be seen to enable one to determine the formational habit. The most striking thing about many of these badly decayed outcrops and cuts is the remarkable way they stand against destruction or removal by ordinary weathering and erosion agents. At many points, road cuts are made, with side walls absolutely vertical, through wholly decayed rock material, that stand for years without crumbling down. Embank- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 35 ments made of earth along roads and trails behave in the same way and one is continually surprised at the steepness of such slopes and their ap- parent stability. Slopes of 40° are not rare on hillsides that are culti- vated, and one occasionally crosses divides that are mere knife edges with slopes of this kind on both sides. Such stability of the soil is a great factor in preserving the agricultural productivity of Porto Rico and in the distribution of its agricultural industries over so much of the in- terior area. There seem to be three factors of large consequence in this stability of the soil mantle. One is the clinging character of some of the vegetation which tends to bind the soil together; another is the small range of tem- perature variation which reduces disintegration or disruption tendencies to a minimum; and still another is the low content of inert or refractory materials, such as quartz, in the rocks whose destruction has furnished the soils; all of which factors favor the making of especially tenacious soil. Most of the soils are for the same reason exceedingly difficult to cultivate. As a direct consequence of this soil behavior and climatic control, there is comparatively little dust formed in Porto Rico. This is especially noticeable on the roads, where one is almost never in the least troubled by it. In two districts, one on the north fringe of El Yunque along the Sabana River and the other on the north side of the divide near Adjuntas, deposits were seen which suggest glacial action. Very large boulders are stranded in positions where it is difficult to account for them by ordinary erosion means, but too little study was given and too little evidence is at hand to warrant a more definite statement. STRUCTURAL FEATURES Most of the structural features represented in the geology are men- tioned in connection with descriptions in other sections of this report. This is especially true of such structures as may be regarded as essen- tially primary. including the interbedded and intrusive forms of various kinds. There is no doubt but that the most prominent structural com- bination in Porto Rican geology is represented by the succession of inter- bedded sedimentary beds and tuffs, cut by or interleaved with intrusive: in the form of sills, dikes or irregular stringers. The combination occurs in great variety as to succession, relative amounts or proportions, quality of material, present condition and relations to other members, but in spite of these variables the structural feature is essentially the same and has the same meaning everywhere. 36 ANNALS NEW YORK ACADEMY OF SCIENCES Igneous Structures The most striking thing in connection with the structure is the re- markable uniformity of the sills and their close resemblance on that ac- count, after weathering, to the fragmental beds with which they are associated. ‘The small amount of metamorphosing influence that they seem to have had, also adds to the difficulty. In some cases, however, a transgressive intrusion has disturbed the adjacent beds a great deal in a mechanical way. In the average case, it is judged that the intruded magma has neither penetrated the materials of the adjacent beds to a noticeable amount, nor Fic. 11.—Shales and ash beds cut by a large irregular dike and sill The dike is shown at the location of the standing figure and the sill extends upward to the left between the plainly bedded layers. Both the dike and the sill are crowded with fragmental material to an extent that makes the intrusive have more the appear- ance of a volcanic fragmental than a true intrusive. has it absorbed or incorporated a great deal of such material. But in a few cases where structural relations were indisputable it was equally clear that the intrusives, both transgressive and concordant, were liter- ally choked with foreign fragmental matter, making them resemble the real tuffs so closely that it is doubtful whether the difference would have been detected except for the clearness of the intrusive relation. Such occurrences suggest that there may well be many other apparently frag- mental interbeds that are in reality fragment-clogged intrusives. On ac- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 374% count of the great abundance of the fragmental matter, it does not seem possible that these intrusive masses could penetrate in that condition to some of the positions where we now find them. It is more likely that a rather fluid magma has penetrated some unusually porous fragmental bed forming a matrix for it, perhaps also spreading it somewhat, and then in breaking across to another bed, in some cases it was still mobile enough to drag the mixture along into the larger transgressive structures. This idea is somewhat supported by the finding of a conglomerate bed, near K-86 on the military road west of Aibonito, impregnated with an igneous matrix in essentially this same manner. In addition to these forms, there are numerous larger intrusive masses, the largest of which deserve a special name. I see no objection to calling them bosses. The two largest occur, one between Caguas and the Caribbean sea toward the south and southeast, and the other between Jayuya and Utuado on the north side of the divide. Volcamc Vent Complexes A special igneous structure that has not been referred to except inci- dentally is that composite of disturbed structures which represents the location of old volcanic vents. They are essentially a complex of irreg- ular intrusive units cutting and including masses and aggregates of various fragmental and sedimentary types in a mixed relation. In the clearest cases, such a complex suddenly takes the place of a formation of apparent promise of continuity such as a series of sedimentary beds, and after an interval these beds are again found continuing as before. For example, the Coamo limestone formation is abruptly cut off and its place is taken for a mile or more by one of these igneous complexes, the lime- stone continuing on the other side again. The Jacaguas reservoir, just above Juana Diaz, les in one of these old voleanic-vent complexes, occu- pying, however, only a portion of the area. Another such case is repre- sented by the complex cutting the great conglomerate beds on the military road at about K-87—88 west of Aibonito. Another is believed to be represented by the very striking basin-like area crossed by the road between San German and Hormigueros. This one is now represented by a very smooth plain five or six miles across surrounded on all sides by more hilly country. The same conditions are undoubtedly indicated by the extremely complex structures seen on the Descalabrado river two miles below, south of the military road. Some of these mark the sites of ancient craters, clogged or choked with fragmental and intruding ma- terials. 38 ANNALS NEW YORK ACADEMY OF SCIENCES Folding Most of the rock formations representing original bedded types have been more or less tilted or otherwise do not now have their original atti- tude. Those belonging to the vounger series, the Tertiary limestones and shales on the north side of the island, are comparatively little dis- turbed, and in some cases do not have a very different dip in spite of the fact that they have changed very much in position with respect to sea level. On the south side of the island, however, even these later beds are in many places tilted at a higher angle than they had in the beginning Fig. 12.—Overturned fold and crush zone in finely bedded shales on the Jayuya road near the summit of the range and occasionally show high angles and even gentle folds. This condition may be seen on the Jacaguas River, near Juana Diaz, better than at most places, but similar conditions are indicated by the relations at certain points farther west. This condition on the south side of the island indi- cates more extensive and more violent dynamic disturbance on this side, which is further supported by the presence of faults cutting and affecting the Tertiary series on its present inner margin. The older series, the pre-Tertiary, is still more profoundly affected and, in almost all districts, shale and ash beds may be found standing at high angles, in many cases practically vertical, and in occasional instances - BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 39 crumpled and overturned in a most complicated manner. Minor fold structures of this kind are especially prominent in the higher ranges, for example, along the Jayuya road along the divide opposite station K-24 of the Ponce-Arecibo road. At such a place may be found as compli- cated structural features of this kind as is usually present in any folded mountain region. The high angle at which such beds stand at many other points leads to the belief that similar complexities characterize a great many of the districts occupying the higher mountain portion of the island as well as some of the marginal areas, but the great amount of erosion and the very limited exposures at most points tend to obscure some of these details. The complex way in which the igneous intrusive masses cut all of these formations also tends to obscure and modify and further complicate the simpler fold structures so that it is not always possible to properly credit the disturbed attitude. On account also of the ‘fact that the total quantity of injected or intruded materials, including dikes, sills and bosses, is exceedingly large and must have caused exten- sive disturbances by reason of the displacement produced by the occupa- tion by these intrusive masses, it is hkely that much of the observed ab- normal attitude of the bedded rocks may be due to this cause rather than to regional folding of a simpler sort. It is fair to say, however, that a sufficient amount of data is not yet available to draw general conclusions as to the meaning of the fold structures in Porto Rico. The striking thing is that all of the older formations are disturbed and that their position and attitude, even along the margins of the island, indicate that the region affected by these movements was more extensive than the present land area. Faulting There are many evidences of faulting on a small scale, in some of which the displacements can be measured. But in most cases the direct evi- dence lies chiefly in the existence of crush zones, slickensided walls and abrupt changes of rock type; there is no opportunity, on account of the general rock complexity, to secure quantitative data. Judging from the difficulty in tracing certain formations between districts where they have been identified, it is probable that there are occasional faults of large dis- placement. Numerous crush zones were seen on the Comerio road espe- cially, but in this case also no system was discernible from the few meas- urements available. The most prominent fault, in its effect upon present features, is the one now marking the inner margin of the younger series of chalky lime- stones and shales constituting the coastal belt along the south side of the 40 ANNALS NEW YORK ACADEMY OF SCIENCES island from Juana Diaz past Ponce at a short distance to the north, eross- ing the Ponce-Arecibo road at K-4.8, and thence westware, crossing the Ponce-Penuelas road at K-10. This is the only large fault actually ob- served that is necessarily of recent age, although a few others are inferred. C abruptly cut off by it. The older rocks of the pre-Tertiary are lifted with respect to the younger series forming the present coastal margin wherever this fault has been seen. It has been traced by us from Juana Diaz to the vicinity of Penelas, a distance of about 12 miles. What becomes of Fic. 13.—Crumpled shales as seen along the Jayuya road near the summit of the range it at either end is not yet determined, but it is believed to extend much farther in both directions. The physiographic habit of the island as a whole tends to support the view that the fundamental structural form is that of a large fault block, with the principal fault displacement and uplift along or near the south- erly margin, tilting the whole mass gently northward. If this disturb- ance took place. as seems to be indicated by the fault described, in very late Tertiary time, accompanying the emergence from the sea, it would account for the abnormalities of Tertiary rock distribution as well as the unsymmetrical position of the main drainage divide. Im any case, how- ever, the fault block structure is a very late development and is superim- posed on the other more complex and older structures of the mass. BERKLEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 44 The island is comparatively abruptly terminated at both the east and west ends. ‘The younger limestone margin, which is fairly continuous along the north coast and extends along about half of the south coast, is wholly wanting at the east end and is also absent at the west end except at the corners. At Fajardo, at the east end and at Rincon, at the west end, for example, the older complex bedded rocks continue to the shore line. In the uplifting of the present island mass, it would therefore ap- pear that breaks occurred at both ends. The included mass is therefore probably bounded roughly on three sides by faults, the east, west and south, and is as a block tilted gently to the north. That there is, besides, considerable differential movement accompany- ing the uplift and disturbance, is indicated by the warping of the erosion plain lying beneath the younger series, the Eocene peneplain, which stands essentially at sea level in the vicinity of Loiza and more than a thousand feet above it at Lares. This difference is accompanied by a much wider belt of these later limestones also in the region about Lares than elsewhere. Such warping need not of course be confined to the last movement; it may have accompanied the depression in eatly Tertiary time, permitting, as is indicated by deposits, very marked differences in the development and encroachment of the organic accumulations. Large Structural Groups Where rock formations or field units are so numerous and so closely related and so complex in primary structure if taken in detail, it is ad- visable to combine them into fewer more generalized groups. A first step of a very general sort, but im all respects sound, has been taken in recog- nizing and using the terms “Younger Series” and “Older Series” in this paper. An additional step has been suggested in recognizing certain smaller associations under the terms San Juan Formation, Arecibo For- mation, Coamo Limestone Formation, Juana Diaz shales and marls, Fajardo shales, Sierra de Cayey tuffs, Ponce chalky limestone, etc., but these are for the most part local designations, some of which may well be expected to become unnecessary after complete correlation is established. A good structural basis for sound subdivision of this sort is not yet worked out. Unconformities There is only one marked unconformity in the structure of the island. This is between the younger and the older series. It measures the break in the sedimentary succession represented by the erosion interval during which this mountain mass, now represented by Porto Rico, was reduced 42 ANNALS NEW YORK ACADEMY OF SCIENCES to a comparatively monotonous surface for the most part at least near to the sea level. The time interval need not have been of very great geo- logic value, but it represents the time between the last violent outbursts of volcanic eruptive activity. occurring near the close of the Cretaceous, and the beginning of simple sedimentation and limestone reef develop- ment and other organic accumulations in the early Tertiary. This un- conformity is very pronounced along the northern margin of the island wherever the two principal series of formations are well developed. This is not easily seen on the south side, but the relative complexities of atti- tude of the two series, together with their position, emphasizes the same fundamental relation. This break in sedimentation is not anywhere marked by a development of a basal conglomerate. In some places the new series is Inaugurated by the development of shales, part of which are hgnitic, indicating land conditions, but in other places such beds are entirely wanting and the upturned eroded members of the older series are followed abruptly by limestones of the reef type. The first type of succession is illustrated in the vicinity of San Sebastian and Lares and the latter type of abrupt limestone succession by the conditions seen on the Arecibo River. It is entirely lhkely that the time value of this break is not everywhere the same. Probably the districts in which shale beds are developed saw the beginning of sedimentary deposition at an earlier period than those in which the shales are entirely lacking. It will be pos- sible to work out these historical and structural differences with further study of the content and distribution. A very extensive development of conglomerates in the region immedi- ately west of Aibonito and smaller occurrences at several other points, especially on the Comerio road south of Bayamon and also near the military road in the vicinity of La Muda, have a suggestion of the possi- bility of a rather important structural break, but there is thus far no conclusive evidence of the presence of any real unconformity. Veins Quartz veins are not prominently developed. There are occasional quartz stringers and in a few places they are abundant enough to make a sort of net work, but in no instance was a large persistent fissure vein seen. In some cases the stringers of quartz carry values in gold, and in all probability they are the source of the placer gold known to exist in certain districts. So far as observations have been made, there is no par- ticular system represented in the vein occurrences. Questions connected with this subject together with mineralization and possible value as mineral resources should be made a special study. BIRKEY, GHOLOGICAL RECONNOISSANCE OF PORTO RICO 48 Minor Structures Although there is extensive development of sedimentary formations which have been subjected to much disturbance, there is comparatively little structure of a minor sort that seems to deserve such discussion in this description. ‘Two, however, that seem to have special significance connected with the origin of the particular beds which have been found are (a) a peculiar crumpled, enterolithic structure seen in one of the ash beds near Guayama, and (0b) the wind-assorted cross-bedding structure of the old dune sands of the San Juan formation at Arecibo. Enterolithic Structure-——The enterolithic structure noted in the ash beds, on account of the thinness of the bed,—about eighteen inches,—and the simplicity of the associated structure,—simple tilted beds,—leads one to believe that the structure is essentially primary rather than of subse- quent dynamic origin. Its appearance is perfectly consistent with the explanation that it is preserved from the time of deposition and its be- havior at that time as a small mud flow. It is a structure such as might be formed by slumping movement of a soft layer. It should be expected that there would be frequent behavior of this kind in the accumulation of such extensive beds of ashy materials, which must in some cases have been deposited under conditions that would make slumping movements possible, but it is not to be expected that material of this kind would in most cases be capable of preserving any of these primary movements. In the case noted, the quality of the interior makeup of the bed seems to have been more favorable to such preservation. It is the only case where such an observation was made. Double Cross-bedding—The cross-bedding structure belonging to the San Juan formation is a prominent feature wherever these ancient dune sands are preserved. A great prevalence of steeply inclined minor struc- tures is crossed by fewer nearly horizontal ones. Measurements made on sea-cliffs a short distance west of Arecibo, where this rock is very promi- nently developed, gave dips of 30 to 33 degrees repeatedly. A series of these is abruptly terminated by a more nearly horizontal bedding for a comparatively short distance and the whole structure is repeated. The layers with this kind of structure are prevailingly one-half to two feet thick and no ripple marks were seen on any of the beds examined. The eross-bedding structure in this case dips always to the west or southwest, and the average strike of the principal beds is about north 30 degrees west. This is consistent with a wind direction not very different from the pres- ent prevailing winds. Very strong structural development of this kind is also to be seen in the city of San Juan at the promontory on which the AA ANNALS NEW YORK ACADEMY OF SCIENCES Morro is built, but measurements of orientation were not taken there. An occurrence of this rock immediately to the east of Arecibo, a short distance south of the lighthouse, showed structures of this kind on a much larger scale than was seen elsewhere, and with an especially imter- esting combination structure. The principal or stronger divisions are widely separated and le nearly horizontal. A comparatively small bed lying in this position was almost unconsolidated, but those strongly cross- bedded immediately above as well as those below were compact enough to stand in a vertical cliff 30 to 40 feet high. The chief interest attaches Fic. 14.—Cliff of the San Juan formation south of the lighthouse at Arecibo The prominent cross-bedding, extending throughout the upper thirty feet of the cliff, is well shown, together with a less strongly marked horizontal structure crossing the same beds. The prominent break near the base is made by a layer of sand which is very poorly consolidated. to the strongly cross-bedded portion forming the upper twenty feet or more of the exposed cliff. The cross-bedding structure itself extends without interruption through a much greater vertical range than in any other outcrop examined, but its attitude and dip were not markedly dif- ferent; the feature that was strikingly different from the structure seen anywhere else was introduced by the presence of less pronounced but still very plainly marked horizontal structures, making an interpenetrating mesh-like arrangement in the face of the cliff. This can be seen strongly enough to show in a photograph even at a distance of 100 feet. It is evi- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 45 dent that some peculiar condition has been in control in the development of this compound structure. There is no reason to assume any difference of source or origin for the steeply inclined structures, or cross-bedding structure, from that assumed for all of the other occurrences of the San Juan formation. It is apparently a perfectly normal product of the as- sorting action and deposition of material under the work of the wind. But under normal conditions it would not happen that a second structure in a horizontal position should be repeatedly developed crossing the well marked cross-beds so that the whole complex combination should be de- veloped on such a scale as is seen in this case. From the nature of the de- posit and the conditions in which it may well be assumed to have been formed—that is, at or near sea level in the vicinity of or bordering upon standing bodies of water—one would be inclined to favor the explanation that the accumulating cross-bedded sands fell into or rolled into a body of water which had a tendency to attack the newly deposited material and to bind the grains together. The difficulty with such an explanation is in the fact that the horizontal structure is repeated at small intervals practically throughout the deposit, and apparently without disturbing the primary depositional structure at all. It would seem quite unlikely - that loose matters of this kind, falling into or rolling into a body of water a. of an open surface sort, should maintain or preserve the primary struc- ture so well. Perhaps it is more logical, in view of all the features, to connect the development of this structure which seems, from its slight influence or modification of the cross-bedding, to be wholly secondary with the subsidence of the coast which is shown to have been one of the late events in the geological history. As subsidence progressed, perhaps somewhat irregularly, it would happen that the ground water level would rise correspondingly high in beds that were passing below sea level. At the surface of this ground water level the tendency would be to accomplish a binding of the granular materials together. Both above and below the ground water level there would probably not be so strong a tendency to develop this binding. With the next step im the progress of subsidence, another streak or indurated zone would be established and these have been repeated throughout the whole formation during the whole period of subsidence. An action of this kind would not tend to disturb the pri- mary structure at all; it would on the contrary tend to preserve it or make it less destructible because of the improved induration. But it could, in addition, develop a succession of secondary structures through- out the whole mass which, if the binding is fairly substantial, might rival the primary structure in prominence when exposed to subsequent de- structive attack. It is possible that such a succession of horizontal struc- 46 ANNALS NEW YORK ACADEMY OF SCIENCES tures could be developed even under a perfectly continuous but very slow subsidence movement by reason of the natural seasonal ground-water fluctuation. From this point of view, the range between two succeeding Fic. 15.—Detail of the double structure in the San Juan formation at Arecibo This photograph was taken at the same point as the one shown in Figure 14 to bring out the horizontal structure crossing the inclined layers. There is no doubt whatever that the dark layer of less consolidated sand in the lower third of the photograph is a primary bedding structure, but the horizontal marks crossing the inclined layers in the upper part of the view are believed to be of secondary origin. BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 4% horizontal markings would measure the fluctuation range of the ground water, the harder zones representing in that case the more persistent upper level for each succeeding depression position. The cross-bedding structure shows as plainly as it does on exposed sur- faces also because of the fact that certain streaks are more perfectly in- durated than the intervening ones, and the objection might well be raised that a history of the kind suggested above would not be expected to de- velop such a difference of induration in layers inclined at such high angles to the horizontal. As a matter of fact, however, these sands are not simple in their makeup. They are in large part fragments of organic material and complete shells of small organisms of a calcareous nature and the primary cross-bedding structure represents an assorting action on this mixture of silicate and carbonate mineral material. It so hap- pens, therefore, that the successive structural units are not necessarily of the same mineral proportions, and in the process of induration, or of binding the grains together, certain streaks yield more readily to this influence and develop greater solidity and resistance to destruction. It happens, furthermore, from a variety of rather unusual primary condi- tions and secondary influences that both a primary and secondary struc- ture of unusual prominence and peculiar association are developed in the San Juan formation. SPECIAL RELIEF FEATURES Playas The flat areas along the coastal margin which are known as playas are all developed at the mouths of rivers and are essentially alluvial deposits of floodplain and delta type. In most cases they seem to occupy areas that must formerly have been embayments in the coast. This develop- ment is most striking, for example, at the mouth of the Arecibo and of the La Plata and Loiza rivers. In some cases, however, there is no marked embayment and the deposit is more strictly marginal, such, for example, as the Fajardo Playa at the east end of the island and others on the south coast. Promontories In addition to the embayments and playas, there are, occasionally, in the intervening spaces, promontories where the rock formations extend into the sea and terminate in cliff forms. These are neither numerous nor are they confined to any particular portion of the island or to any rock formation. They are represented by the most recent of all of the 48 ANNALS NEW YORK ACADEMY OF SCIENCES formations, essentially a silicified dune deposit such as that at San Juan, also by Tertiary limestone reefs, such-as that at Quebradillas and at Guanica, or by the still older igneous and clastic series, such as that at Anasco, or by massive intrusives of a strictly igneous habit, such as that at Maunabo. It would appear from this that the former outline of the island must have been more irregular than it is at present and that the distribution of marginal formations is also not as regular as has been represented in earlier reports. Fic. 16.—Playa plain and marginal terrace View of the Playa plain (foreground), the marginal terrace (middle field) and the mountainous divide formed by the Sierra de Cayey as seen from the ‘Central Machete’ near Guayama. This terrace bevels across the upturned edges of shales, ashes and intru- Sives of the older series and is probably due to marine cutting. Terraces At many places on both sides of the island there are comparatively smooth tracts having the appearance of bordering shelves which represent true terraces. Their location along the sea margin and the compara- tively insignificant development of similar benches along the streams lead one to believe that they have an origin connected with the wave action and attack of the sea.. This interpretation is supported by the presence of these terraces along the coastal margin where stream action would not seem to have been able to reach. In any case, the presence of such terraces, which stand from 100 to 200 feet above the present sea level, indicate a former more submerged condition, so that the sea or streams, or both combined, were able to attempt base-leveling at that ele- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 49 vation. The presence of great quantities of roughly assorted gravels clogging some of the valleys of the southerly side of the island tends to support the same general conclusion. The bearing of these features on the geological history of Porto Rico will be taken up at another point. Cuestas Both on the north side, for nearly the whole length of the island, and on the south side, over the westerly half, there is a bordering belt of limestone and associated beds that have been developed on an eroded sur- Fig. 17.—Structure beneath the marginal terraces Strongly bedded ash together with associated shales cut by small dikes forming a part of the terrace near Guayama. These rocks belong to the older series and dip into or toward the mountains rather than toward the sea. face which beveled across the more complex structures of the older series of formations that formerly made up the mass of the island. These limestone beds are several hundred feet in thickness and dip gently toward the sea. On the inner margin of their present extent toward the interior, especially along the north side of the island, they are abruptly terminated in a very irregular line of modified cliff forms facing toward the prevailingly smoother and lower ground for some distance toward the interior. For the most part, this limestone margin is exceedingly rugged and broken. The width of the belt with this rugged character 50 ANNALS NEW YORK ACADEMY OF SCIENCES varies very much in different parts of the field. Its most striking devel- opment is in the district extending from Tao Alto to Aguadilla. In the district extending eastward from San Juan and also in some of the areas on the south side, this margin is very much broken and so obscure in some parts as to escape detection. In its best development. however, it is a typical cuesta, formed in the usual manner by the erosion of a for- mation representing a recently uplifted coastal series: ‘The series of formations involved formerly extended inland very much farther than they do now. Only the outer margin remains from the erosional de- struction of a series of beds and reefs that in former times covered a Fig. 18.—IJnner lowland near Bayamon View looking north from the Bayamon-Comerio road toward San Juan, showing the monotonous features of the lowland belt in the foreground and the comparatively promi- nent hill remnants of the Tertiary formation cuesta in the background. large portion of the island. -The road running from Aguadilla to Moca, San Sebastian and Lares extends for practically the whole distance, after leaving the coast, along the inner lowland at the foot of this cuesta or along the cliff forming the inface. The same features characterize the surface topography as far east as Corozal. This feature is much less pronounced on the south side of the island. Peneplain Beneath the limestones constituting the cuesta and representing the Tertiary series there are, in numerous places, traces of a former plain that represented the results of erosion on rocks that had a complex struc- BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 51 ture. Occasional profiles of more distant ridges also show a sky line that suggests the former existence of such a plain, and in favorable localities it can be traced directly to the foot of the limestone cuesta. Occasional traces are also seen on more mountainous tracts, especially at the west end of the island, near Rincon and in the vicinity of Mayaguez. At the latter place, these remnants of the old plain are called mesas. It is the judgment of the writer that these all belong to a single base-leveling sur- face or marine-cut platform formed in the period just preceding the development of the Tertiary limestone series. Judging from beds devel- oped immediately upon this surface, it must have been completed in early Fig. 19.—Haystack (pepino) hills A characteristic view, showing the small soil-covered flats and associated haystack hills found in the region of the Arecibo formation. Photograph taken on the road be- tween Arecibo and Barceloneta. Kocene time, and perhaps was even largely developed in pre-Tertiary time. It may be referred to as the early Tertiary base level or peneplain. There are many minor features giving variety to the surface relief which depend for their particular relations and character on underground structures which are as yet imperfectly understood. Haystack Hills The most striking topographic feature of the whole island is the re- markable development of small isolated or grouped rugged hills usually rising abruptly above adjacent smooth flat soil-covered areas at various levels throughout a broad marginal belt along the north coast, west of ~ 59 ANNALS NEW YORK ACADEMY OF SCIENCES San Juan. They constitute a feature so unusual that even the un- trained casual visitor is impressed with them. This feature has been referred to before in connection with the de- scription of the “Younger series” of rocks, especially the Arecibo reef limestone formation. In spite of the unusual appearance presented by this distribution of “haystack” hills and intervening flats, their origin is judged to be comparatively simple. The active agents and processes have been the same as those at work on all other parts of the island, but the results differ because of the fundamental difference of material and Fig. 20.—Care structure in the haystack (pepino) hills Near view of the limestone hills forming the margins of the small cultivated flats in the typical haystack hills district. This view shows the cavernous nature of the lime- stone forming these hills, a structure that is regarded as the most significant feature and probably the largest factor in the development of these peculiar relief forms. structure. Nowhere are these features developed except where the later reef limestones are the underlying bed-rock formation. The essential steps in the development of these forms are the follow- ing: The reef limestones are not uniform in composition or structure. They have more or less intermixture of earthy matters which are distrib- uted irregularly, but chiefly at certain horizons, as more earthy or shaly beds of no very great lateral or vertical extent. Such conditions are re- BERKEY, GHOLOGICAL RECONNOISSANCE OF PORTO RICO 58 peated at occasional intervals in successive horizons. As such a series is lifted above sea level and subjected to ordinary erosion and weathering, the tendency is, (a) for the purer and more massive reef limestones to be attacked by the solvent action of percolating water with a development of underground channels, porous rock condition and actual caverns, (b) for the more earthy layers to resist and limit such action at the levels where this matter is present in sufficient abundance, with a development of residuary material. As this action progresses toward maturity, many of the larger caves collapse and sink holes are thus formed. With still fur- ther development, the sink holes merge into each other in local areas where solution has been most active, the earthy debris forms a soil in the bottom corresponding in level with the first important earthy layer, and adjacent remnants of the limestone reef stand out as sharp rugged hills separated by irregular notches that represent other smaller collapsed eaves. The result of such action and conditions, finally, is the numerous “haystack” hills standing on flat soil-covered areas or surrounding such areas as if they were just set down as bunches on this surface. This re- lation is repeated at different levels throughout the belt from San Juan to Aguadilla, but the most striking developments are local, apparently where the structural relations are just right, and may be seen best be- tween Tao Alto and Arecibo, especially in the vicinity of Manati and Vega Alta. It was at first thought that former subsidence levels might have some- thing to do with establishing the level tracts, but the observation that these tracts stand at. very different levels in immediately adjacent dis- tricts together with recognition of the structural difference, lead us to give credit to the primary structural character of the formation itself as the controlling factor in the present distribution. According to this ex- planation, these hills are mere remnants left from solution attack on a reef limestone, the depressions between them representing collapsed cay- erns, the walls of which may still be seen on the sides of some of the more rugged hills, and the surrounding or intervening tracts are soil-covered and level, chiefly because of the accumulation of earthy material, left be- hind after removal of the overlying reef, now halted in its reduction at the first important less soluble beds. MINERAL RESOURCES © An examination of specimens of minerals and ores in the hands of local prospectors and residents interested in developing mineral resources, ®A good list or tabulation of the mineral occurrences of Porto Rico may be found in the article by H. C. B. Nitze listed at the close of this paper. 54. ANNALS NEW YORK ACADEMY OF SCIENCES together with observations made personally, shows that there is consider- able range of minerals and ores. It appears also that considerable atten- tion has been given in a few cases to local development. ‘There is large variety shown in a collection of this material and in some cases the speci- mens exhibited look very promising indeed. But there is almost no reliable information touching the quantity or the exact relations or esti- mates of possible profitable development. It can be said, without danger of contradiction, that none of the developments so far undertaken looking toward the systematic mining have proven profitable. Gold Only one enterprise of this kind seems to furnish any production, and this is the placer mining for gold. Gold washing has been practiced from the early Spanish occupancy to the present time, and it is not at all a rare thing to see several men digging in the stream gravels for the “nay dirt” and panning out the gold. This is done in all cases on a very small scale and with the aid of the simplest equipment, and the returns appear to be very moderate. It is claimed that in former times a much more elaborate system of working such deposits was in operation under the Spanish regime, and, according to historical statements, they were ‘ considered profitable. More recently, there has been at least one attempt near Corozal to develop this kind of ground by the use of modern ap- pliances, but the plant has been allowed to go to entire ruin. The only places where actual placer washing was seen in progress was three miles south of Corozal and on the Sabana river near Luquillo. Near Corozal, also, some work has been done in an attempt to discover the veins or lode which may have furnished the placer gold. There are several pits, trenches and shafts, in some of which quartzose stringers were seen which appear to fulfill the requirements of a source of supply. Some free gold was found in panning a little of the weathered material at one of these spots. There is little doubt but that these veinlets or stringers, which were numerous at one of the cuts, are in part the sources of the placer gold of this locahty. But at no place examined was there to be seen any “vein” of apparent consequence or any structure suggesting the course or extent of the mineralization. Of course the rather mixed state repre- sented by the residuary matter, seen almost everywhere at the surface, does not lend itself readily to the tracing of veins, and it may therefore happen that conditions would prove, after thorough exploration, to be better than the first brief examination indicated. There are said to be some old abandoned workings dating back to Spanish conquest times at ff BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 55 the same locality, but such evidences are very obscure and would seem at best to have very little bearing on present prospects. Copper Some very good specimens of copper ores were seen in the possession of Mr. Henry D. Sayre, of Corozal, who assured us that there were several localities represented in the collection and that in no case had the real value of the occurrence been determined. It was understood that exami- nations have been made by engineers sent to Porto Rico in the interest of American mining companies, and that some exploratory work has been carried on by. Porto Rican companies or groups of individuals, but so far as known there is at the present time no development work being con- ducted, and the possible value of these deposits has not been thoroughly proven. Zinc, Lead and Silver One prospect, on which several hundred feet of underground work has been done, was visited at Barrio del Carme in the Sierra de Cayey, on land owned by Pablo Vasques, several miles northwest of Guayama. The country rock is chiefly andesitic tuffs cut by porphyritic intrusives. A quartz vein carrying sulphides, pyrites, sphalerite, galenite and chalcopy- rite has been followed and there is some ore on the dump. ‘The vein varies from a mere streak to a width of two feet. The mineralization is irregular and the values are said to be chiefly in lead and silver. The second-class ore is essentially mineralized tuff. The exploratory work has been done in large part on side slips and streaks quite outside of the vein proper. In all of this side work, there was apparently no new min- eral-bearing ground discovered. The first-grade ore is heavy and the dis- tribution of values is not determined. There is no doubt of the existence of a real vein or of the ore in this case, but there is need of more intelli- _ gent exploratory development along the vein proper, together with a study of the possibilities of separating the chief values by some sort of milling operation, before a reliable conclusion could be reached as to pos- sibility of working the deposit as a mine. Iron - One magnetic iron prospect was visited. This occurrence is on the divide about ten kilometers west of Naguabo. It is reached by driving eut on the road from Naguabo toward Torres to about this distance and then taking saddle horses to the divide, a distance of about two kilometers southward. On the expedition we were accompanied by Mr. Arturo Gallardo, Jr., Alealde Municipal of Naguabo. 56 ANNALS NEW YORK ACADEMY OF SCIENCES There are many surface bowlders of magnetic iron of fine quality in this vicinity. A lhttle underground working is evident at one point but this is now caved in. Surface observations, together with a few magnetic observations, failed to show any very extensive deposit at that point, but the quality appears to be good in iron content. The ore carries a little copper and is associated with an igneous rock essentially andesitic in composition. It could be traced with a fair degree of certainty about fifty feet east and west just below the crest of the ridge. Considering the associations at this place, it seems necessary to conclude that the ore is igneous in origin and that it probably accompanies one of the porphyrite intrusives. Other occurrences of similar ores were mentioned to us in this same region, but none were visited. Coal and O1l There is no good ground for believing that valuable resources of these products exist in Porto Rico. Some prospecting for them is carried on, however, in a desultory way. The only basis for the hope of finding coal is the occurrence of lignite and lignitic material with the shales lying at the base of the younger series of rock formations, below the Arecibo lime- stone member. JLignitic material was seen in these shales near Lares, and similar or better material has been reported from near San Sebas- tian. From what has been seen, there seems to be no promise of very valuable deposits of this kind. The structure is simple and a very little exploratory work done in a systematic manner would determine the prob- able value of every occurrence known. There is no promise at all of such content in the older series. No oil indications were observed. The only formation to be considered in investigating the prospect of oil is also the basal shales of the younger series. Limerock A particularly porous, granular and uniform limerock is obtained from the small island, Icacos, just off the northeast coast, and is used in sugar refining at the Central, owned by the Bird brothers, at Fajardo. The rock is organic, largely foraminiferal, and is probably structurally of the same origin as the San Juan dune sand deposits,—comparatively recent. Such materials are doubtless to be found in large amount, but not always so pure and so uniform in quality and structure. Limestone suitable for lime burning or for cement mixture is certainly not rare. Limestone of a quality that would permit its use in structural work is also found at some places, but apparently very little native stone is used. BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO. 57 Guano Bat guano is found in some of the caves in sufficient amount to be a source of local fertilizer. Road Metal Several kinds of stone were seen used extensively in road improve- ments. The particular variety used depends largely on the local supply, but the most common are the Arecibo limestone and the massive syenite and granite porphyry. There is no lack of these as well as other types suitable for such use. There are other mineral substances that will invite investigation, but no-others came under the writer’s personal observation and no facts re- garding them are in hand. HOT SPRINGS Thermal springs are known in the vicinity of Coamo. Judging from their location and apparent relation to other physical features, they are believed to lie along a fault weakness. The district is also one of com- paratively late igneous activity, and this, coupled with the other factors, leads to the suspicion that the springs are directly connected with the dying igneous activity and may actually represent juvenile waters. At Coamo Springs Hotel, one of these springs has been developed and controlled for commercial purposes. The water is hot as it comes out of the side of a small ravine and runs down over the slope, which is covered to a moderate extent with deposit from these waters. The immediate bed rock is not well exposed, but it is judged to be either a tuff or an igneous complex and the field relations in the vicinity show that there is a thick series of beds both above and below. No doubt critical field study could determine the actual relations with considerable certainty. An analysis of these waters, furnished by the proprietor of the springs and made at the agricultural experiment station at Mayaguez, is as fol- lows: Fixed elements per litre of water Biber sania seit... Beto l coe ae ae ek eek 0.01296 RE OTe HIN Mor s= semie ote ts. ct alt cb, creole Late 0.79902 SSELFPLODESSLT ESOC eae ae a a ee a en 0.52531 NCEA AOL ASST: oe oe es air ebm Scie stile even ewes se O.Q0082 UUM GE SOIT co ccbak fcc vey sce cdua Gace esbece’e 0.23054 0 LEE SUT 1 ERS AE ee a ee 0.08127 LSP IGT SUPES U1 ONS 00 hapa 2 ne Oe ee 0.03503 Carbonate of iron........ See hale d ea tciors ou Wd dete S Ab ace 0.01114 ee eas acre na) Sib. oie so Re hae view ee es 1.68559 58 ANNALS NEW YORK ACADEMY OF SCIENCES Gases in solution per litre of water at 0° of temperature and 760 mm. of pressure Nitrogen: -. 223 «3 oes oie Oe a eee eee 13 ce. 740 OXY SOB. 2255 kn FRE Stee es oe he Ree ok ee 761 Sulphyd@ric Reise cee see ae ee ae ers 1 967 Potal Fo kas 2 eae ek tee 15 2468 - HistTorRIcAL STATEMENT A complete or even a reasonably full account of the geological history of Porto Rico cannot be written at this stage of the investigations. Such a statement is necessarily the end product or climax of the whole series of studies that are proposed, but it may not be out of place to outline some of the leading and most clearly marked steps as a rough sketch or a preliminary attempt. At the outset, it is well tc appreciate that the Island of Porto Rico is geologically young. There are no traces, so far as known, of any of the so-called ancient rocks. It is quite true, of course, that the older series of formations is largely a volcanic complex whose exact age may never be - accurately determined, but there is no occurrence of profoundly meta- morphosed members or other evidences of great geologic age. Besides, the series, complex as it is and difficult to group into suitable divisions as it may be, undoubtedly forms a very closely related succession of minor formational units whose uppermost members are determinable as to age with reasonable accuracy. It would appear also from the nature of the deposits and their structural relations that the accumulation must have been, for the most part, a rapid process. There is no good reason, so far as any of these facts are concerned, why the whole of the “Older series” could not have been accumulated in a single geologic period. ‘The fossil content of the upper members of this series indicates that this period was the Cretaceous, as used in the broader sense in geology. Whether or not the older members date back to an im- mediately preceding time cannot yet be definitely stated, but whatever there is, is clearly so closely associated with the Cretaceous beds that they can all be treated as a single historical unit. This earlier period is characterized by volcanic and other igneous activ- ity on a very large scale. Beds were accumulated both above and below sea level. There seem to have been oscillations of level accompanied by recurrences of similar beds, and apparently much shifting of the supply of materia] accompanied by great variation of character laterally. There is good evidence that succeeding volcanic outbursts broke through these beds at many places. BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 59 An occasional more prominent change of conditions, more or less clearly marked in the structural relations and character of material, may possibly be used as a basis for epochal subdivision. It is quite clear, how- ever, that there was no profound change of geologic control throughout the whole of this earlier time,—-it was strictly a volcanic period. The succession of disturbances by which it was affected is represented in part by dynamic modifications of the nature of folding, crushing and faulting, but this is probably an accompaniment of the more profound igneous activities also, and need not be regarded as evidence of any strikingly different causal process. If there were contributory causes of a regional sort controlling the folding, they are essentially simply superimposed upon or introduced into the larger, more profound and longer continued igneous activities which prevailed both before and after that time. This long period, characterized by great complexity of formational development, including tuffs, agglomerates, conglomerates, shales, lime- ~ stones and immense numbers of intrusives of great variation in size, form and composition, finally came to an end by the dying out of the volcanic energy, and greater stability of the whole with respect to elevation and subsidence was established. Erosion cut down the exposed formations, the sea attacked the margins and in time most of the projecting moun- tain mass was reduced near to base level, the sea encroached far onto the former land areas and a new historical chapter was begun. It is not possible to say, with the data in hand, that the entire island was reduced to a peneplain, or perhaps a conoplain, but there is good evi- dence, from the traces still left of former planation and from the dispo- sition of the remnants of overlying formations still preserved, that the greater part of the present area was worn down to base level and sub- merged. ‘The process of base leveling was going on before the close of igneous activity and it was continued long enough to bevel across rocks of all sorts with marked success, but there is no necessity for regarding it as a very long geological time. As erosion proceeded, sediments were deposited unconformably around the margins of the island of that time and perhaps also in some of the marginal valleys above sea level. These constitute the earliest shale beds of the “Younger series” and are believed to be of Eocene age. They are at least early Tertiary. Where more simple marine conditions came into control, as would happen when submergence or planation had masked or destroyed the more elevated sources of supply, the deposits became almost wholly reef limestones and shell limestones, with only minor amounts of strictly detrital material irregularly distributed. This gave a succession of somewhat irregular beds which are abundantly supplied with organic 60 ANNALS NEW YORK ACADEMY OF SCIENCES remains and which bear evidence of the continued depression favorable for the growth of these accumulations for a considerable part of Tertiary time. ‘There is some suggestion in the relations shown in the eastern portion of the island that this end was not wholly submerged and that differential subsidence gave to this portion less prominent development of the heavy, massive limestone beds. In later Tertiary time there was marked reémergence from the sea, accompanied by warping, so that the later limestones and reefs were lifted to very different elevations in different parts of the island margin. Since that event, the whole has been again subjected to erosional attack of the sea, and to wind work, with the result as now seen in the physical feat- ures. The comparatively easily destroyed shales, marls and limestones of the Tertiary series have been extensively removed, leaving only a fringe of these formations along the north coast and a part of the dis- tance along the south coast, and developing all of the topographic forms characteristic of the erosion of emerged coastal deposits, together with some very special forms due to the peculiar makeup and attitude of the rocks themselves. Since this first emergence there have been minor oscillations also, the record of which is observable in marginal terraces, deeply trenched flood- plain deposits, and thoroughly indurated wind deposits of presumably Pleistocene age. Apparently the latest movement has been one of slight emergence. A summary, therefore, of the larger items in the geologic history in- cludes the following: - 1) A long geologic period of voleanic activity, accompanied by marginal at- tempts at assorting of fragmental and detrital material and organic accumu- lation disturbed from time to time by renewed or extended igneous activity. 2) A dying out of volcanic energy, greater stability of the mass with respect to elevation and subsidence, and erosional attack continued long enough to re- sult in extended planation and partial base leveling with final extensive sub- mergence. 3) The development -of an unconformable overlying series of shales, reef limestones and related deposits chiefly of organic origin, brought to an end by final re-emergence. 4) The development of present surface features under stream erosion and marine marginal attack, with modifications arising from oscillation of level. The geologic column forming the basis of this outline, avoiding minor details that are properly the subject of further study before specific state- ment should be made, is as follows: BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 61 Recent alluvial deposits. Submergence with flood-plain deposits. Younger Series. San Juan dune sands (Pleistocene). Submergence with terrace cutting. ‘Post-Arecibo emergence and erosion. Organic limestones, marls ete. (Mid-Tertiary). Arecibo reef limestones (Oligocene). Lares shales ete. (Eocene ?). Marked unconformity. Older Series. Interbedded limestones, tuffs and shales etc. (Cretaceous), Coamo tuff-limestone, Trujillo Alto limestone, Aibonito conglomerates and shales with many intrusives. Interbedded foraminiferal shales and ash shales with tuffs, cut by many intrusives (Early Cretaceous ?), Fajardo shales, Mayaguez shales, Barranquitas shaly limestones, Sierra de Cayey tuffs, Comerio consoli- dated ash beds and tuffs, etc. FuTURE PROBLEMS One of the objects of this exploratory study was to discover and define the problems that should be investigated by this organization. It is not supposed, in enumerating this list, that these cover every possible subject of special study, but they do indicate the fields in which there is promise of immediate and valuable scientific returns, and at the same time will add to the fund of usable information to be put within reach of the people of Porto Rico. BASE MAP One of the fundamental things as a basis for all sorts of detailed geo- logic work is a good contour map. The whole island ought to be mapped in the same manner as is done in the United States, using the same quadrangle system. On account of the density of population, the com- plexity of structure and relief and the variety of agricultural uses of the soil, the scale should be approximately one mile to the inch, or 1: 62500, so that these maps could be used as base maps for all sorts of special purposes. The maps now available are chiefly those of the Interior Department of Porto Rico, made to illustrate the various reports of the department and representing the progress of public works such as railways, telegraph and telephone lines and highways. On account of the care with which the different classes of roads have been shown, and the general accuracy of locations, these maps are especially useful in the present investiga- tions. One of the most useful is a map of the Bureau of Public Works 62 ANNALS NEW YORK -ACADEMY OF SCIENCES on a scale of approximately 2 inch per mile which has even the kilometer distances along the roads indicated. Until some sort of a contour map can be secured. such maps as these will be found eminently serviceable. Along the south coastal margin within the region of irrigation devel- opments, there has been some special mapping with contours. In no case do they cover much ground beyond the outer lowland and terrace border, and because of this limitation they are not so generally useful for our purpose as the Interior Department maps. They are, however, very- much more accurate and detailed and for the territory covered are emi- nently suitable as base maps. GEOLOGIC MAP A geologic map of the island should be one of the results of this series of studies, whether a relief map is secured or not. Such a map of the whole island is necessarily an ultimate rather than an immediate product, but district maps can be undertaken at once, with no difficulty whatever. These preliminary districts can be selected so as to include some of the most promising investigation problems in special lines, and both kinds of work can thus be carried on at the same time. This therefore leads directly to the next item, which is district studies. The only geologic map thus far attempied is that by R. T. Hill. DISTRICT STUDIES It is possible now to select areas which are known to contain geologic features of special interest and significance, and it will generally be con- venient, if not indeed necessary, for the investigator to make a detailed geologic map as a secure foundation for his special studies. One of these is the Coamo Springs District, which may be made large enough to ex- tend from the Descalabrado river on the west. to Salinas on the east, and reach as far north as Aibonito. It will include as features of special importance for investigation, in addition to the mapping, the hot springs, the great conglomerate series, one of the later of the great volcanic vent. complexes, the genetic history and horizon of the Coamo limestone which is a striking mixture of volcanic and organic matters, the high floodplam deposits of the stream valleys and their bearing on late geologic history, and certain physiographic studies connected with the coastal terraces. This district promises, as can be seen, an unusually large range of topics - inviting special study, all of which will be illuminating to further devel- opment of the geologic survey of the island. Another district of equal promise in a very different manner is on the north coast extending from the Quebradillas to the Arecibo river and ihe BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 63 reaching from the sea to Lares so as to include a strip of the older com- plex rock series beyond the inner margin of the Tertiary series of reef limestones and shales which constitute the greater part of the bed rock of the area. Beside the mapping and detail of structural relations, this district presents the best opportunity to investigate the question of exact age of the basal beds of the Tertiary series, the transition from lignitic shales of perhaps fresh water alluvial origin to massive limestones of reef type, a subdivision of the Tertiary series, the meaning of the thinning out and disappearance of the Lares shales toward the east, and a paleon- tologic study of the beds, all of which are fundamental in any additional study of other districts containing the Tertiary rocks. There are besides good opportunities to study the meaning of the high terrace-like shelf coming abruptly to the sea at Quebradillas river and the meaning of the deep embayments now occupied by such playas as that at Arecibo. This is also one of the best localities for a detailed study of the structural and petrographic features of the San Juan formation as well as the behavior of modern dune sands along the present coast. Another district of still different features, and giving foundation for special studies of quite a different bearing, is that lying between Caguas and the Caribbean sea and perhaps extending as far eastward as Naguabo. This will include the largest massive igneous unit in the whole island and promises information bearing upon magmatic differentiation, origin of the magnetic iron ores, relation of the great intrusive masses to the other igneous representatives, petrographic range of the igneous rocks, and marginal metamorphic or other effects,—studies fundamental in a final statement of the igneous history of the island. There are other districts which have special problems associated with the regular areal geologic work, but these are sufficient to indicate the range of such district studies and their variety. Certain special investigations are of a sort requiring comparison and summary of many different localities, and for these it will not be wise to handicap the investigator by limiting work to a single district. Some of these are suggested below. | REEF-BUILDING ORGANISMS The limestones of Porto Rico are remarkable for the great prominence of alge and corals and other closely associated organisms lending them- selves to the construction of reefs and accompanying deposits. These forms belong to practically every limestone formation of both the older and the younger series except those most closely related to the shales. It is a study requiring the training of a specialist in such lines. 64. ANNALS NEW YORK ACADEMY OF SCIENCES PALEONTOLOGY The total organic content is much greater than is intended to be in- cluded under “Reef-building organisms.” There are immense numbers of splendidly preserved fossil species of organisms belonging especially to the Tertiary series. Probably a great many are new to science. There are probably few places in America or within territory belonging to the United States where the marine Tertiary succession is of more promise than in Porto Rico. This problem or line of investigation is closely related to the next topic, that of Tertiary subdivision. TERTIARY SUBDIVISION A faunal and structural summary will naturally lead to the establish- ing of subdivisions and the determination of horizons in the younger series of rocks culminating in a statement of the complete Tertiary his- tory of the island. SAN JUAN FORMATION A study of the characteristics and detail of origin and historical steps associated with the Pleistocene fossil dune sands, referred to as the San Juan formation, is another problem. “SUBDIVISION OF PRE-TERTIARY COMPLEX A discussion of this kind is one that will properly follow upon the completion of areal work in several of the typical districts. It is, how- ever, one that will necessitate investigations throughout the interior of the island, and will include a summary of the characteristics of all of the prominent local formations. A grouping and correlation cin no doubt be made in due time. MINERAL RESOURCES On account of the interest taken by the people of Porto Rico in the question of possible mineral resources, it is desirable to undertake an investigation of the kinds of products, their origin, distribution and prob- able economic value. Jn connection with this, because of the small amount of exploratory work that has been done, it would be especially useful if suggestions were made at the same time about the methods of exploratory development and the people cautioned concerning wasteful methods. This work should be done so as to cover the whole range of mineral possibilities in the island regardless of location. ‘There are known deposits of copper, iron, gold, lead, silver and zinc among the ke BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 65 metals as well as a number of non-metallic products. But in no case is the real value, or the probable extent or the geological relation, known sufficiently well at the present time to serve as a basis for a discussion. This should be one of the first undertakings of this survey, both because of the fact that its value is fully appreciated by the people of Porto Rico and because its conclusions do not materially depend upon the other in- vestigations or mapping progress. PETROGRAPHY On account of the great number of igneous rock occurrences and the very great variety that is certain to be shown in quality and minor petrographic character, and because of the considerable range in compo- sition already known to characterize the intrusives, there would seem to be an ample and promising field in this line for a special investigation. It is possible that some genetic relationship is exhibited in the distribu- tion of these variations and that a thorough comprehensive study would throw some light on the more obscure problems connected with the gen- esis of igneous rocks. This is a problem that can be taken up at any time, and that need not be regarded as dependent upon special district studies, although it is evident that the finishing of work on certain dis- tricts would facilitate a study of this kind. PHYSIOGRAPHY Enough is known of the physiographic features and their meaning to appreciate that a great deal of the detail of the later geologic history of the island is more or less intimately bound up in the physiographic de- velopment. The broader or larger physiographic features have already been suggested, but there are certainly many details, some of which may well be of much significance in understanding the geologic history, which will require the special attention of a trained physiographer. Porto Rico is a unit of geologic history, of geologic structure and of physiographic form. Each is of sufficient complexity and unity to be made independent subjects of investigation. THERMAL WATERS The hot springs in the vicinity of Coamo Springs suggest from their situation and reported composition the possibility of being representa- tives of juvenile waters. It is believed that a study, planned especially to investigate the origin and character of these waters, together with such others as may exist, would be a very suitable special investigation. On 66 ANNALS NEW YORK ACADEMY OF SCIENCES account of the fact that the principal occurrence of this type of water is very local, it would be possible to combine a study of this kind with a district study such as has been referred to in a preceding paragraph. GEOLOGIC HISTORY The complete geologic history of the Island of Porto Rico cannot be written until all of these and perhaps other more special investigations have been made. A complete historical statement must be regarded as an end product of the whole range of studies carried out for more special purposes. It is, therefore, the final topic and may well be deferred to, a time when most of these already suggested have been carried far enough so that the data of special importance secured by them are available for this general summary. It is evident from the appearance of this list that there is a very great amount of geological work awaiting the investigator in Porto Rico, and that it is varied and complicated enough to require several years of study in large part by experts or specialists in all branches of the subject. It can be seen also that the Island of Porto Rico is a geographic unit of more than usual complexity and scientific interest and gives promise of results for effort expended in researches along geologic lines. COLLECTIONS A beginning has been made toward securing a representative collection of typical rocks and fossils. Several hundred specimens were brought to New York for use in formulating the accompanying description, and as a possible basis for further more special investigations. As a first step in this direction, about a hundred thin sections of the rocks have been made for microscopic study and detailed comparison. They will form a basis in planning the special petrographic investiga- tions which may be undertaken. In lke manner a large number of fossils have been gathered and their general relations are being studied. Additional investigations along paleontologic lines will be in large part outlined or suggested by the trend of these studies, for although the collection is very fragmentary it is nevertheless characteristic and fairly representative of the principal for- mations. - More than a hundred photographs were taken of strictly geological subjects illustrating typical physiographic features, structural detail of rock formations, structural relations, etc. These are all suitably labeled and form the beginning of a collection of ilustrations of Porto: Riean BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 6% geology. In addition to the regular photographs, a number of photomi- crographs have been made from the thin sections of typical rocks. Some of this material will finally serve as a foundation for the geo- logical section of what it is hoped may become a Natural History Museum of Porto Rico. ILLUSTRATIONS CROSS-SECTIONS The note books of the party and field maps carry a record of field de- terminations and detail of structural relations and comments greatly in excess of what can be published in such a report. They are the property of the organization and are of particular service as guides in planning further work and in giving each new investigator his bearings, together with some suggestions about the character of his own district or the dis- tribution of data bearing upon his special investigation. The note books contain observations along some of the principal roads in sufficient detail to serve as a foundation for complete generalized geologic cross-sections of the island on two especially important lines. Cross-sections, therefore, have been drawn to illustrate the kind of sur- face relief, the grade of the road, the kinds of rocks or rock formations and the geologic structural relations, and are reproduced to accompany this report. An immense amount of detail is necessarily omitted or com- bined into generalizations in order to bring the sections within the scope of a publication of this kind. It is judged that some of these details will be suitable illustrative matter for future reports based on studies of spe- cial districts. One of the sections is based on data gathered along the road from Ponce to Arecibo. The line is drawn from Arecibo to Ponce direct and the data are projected to this line. This method tends to obliterate the windings of the road and secure practically normal propor- tions and relative positions for the associated formational units. ‘The other section line is drawn directly from San Juan Point to Santa Isabel. By projecting to this line all the data gathered on the Bayamon-Comerio- Barranquitas-Coamo road a great deal more elimination of road curves is accomplished than in the other section, and it makes the grades of the road look somewhat abnormal by reason of this shortening of road dis- tance of certain large curves, but on the whole the relations are shown without special difficulty except that attendant upon the need of general- izing the minor structural detail. 68 ANNALS NEW YORK ACADEMY OF SCIENCES MAPS A hasty reconnoissance examination is seldom a satisfactory basis for an areal map. ‘This is quite true of the present investigation.- On this account, therefore, if it were not for other considerations, an areal map would not be attempted. But in this case, where a good many more or less independent special investigations are to be carried on in which a reasonably accurate geological map will prove decidedly helpful, there is sufficient excuse for presenting a reconnoissance map. An earlier map of this kind prepared by R. T. Hill was made under conditions so much less favorable for travel, and seems to have been constructed in some par- ticulars with so much less opportunity for observing the actual conditions in certain areas, that an entirely new map is believed to be the better solution of the present need. The accompanying reconnoissance map is intended, therefore, as a convenient guide or location map for subsequent more special investigations, and it is expected to be wholly replaced by one of much more detail and greater accuracy as a final product of this survey. ACKNOWLEDGMENTS The members of this expedition have appreciated the very material help, the sound advice and useful suggestions given by the officials of the government of Porto Rico, and are indebted to Governor Yeager for his very practical aid in making arrangements for the field work and for his live interest in these investigations; to Colonel Shanton, chief of the In- sular Police, for his willingness to give introductions to men acquainted with special mineral localities and for his precautions to insure protec- tion against unnecessary delays; to Mr. Wheeler, of the Interior Depart- ment, for assistance in securing suitable maps as a basis for travel and notes; to Mr. Bonner, the Auditor, for facilitating the settlement of ac- counts; to Dr. Lippitt, of the Bureau of Sanitation, for information regarding sanitary precautions and hotel accommodations; and to Mr. Campbell, of the Bureau of Transportation, for the excellent equipment for travel which contributed largely to the success of the expedition. Many others have been of assistance in pointing out localities of special interest, in giving names of reliable informants and in acquainting us with the usages and customs of the country. BERKEY, GEOLOGICAL RECONNOISSANCE OF PORTO RICO 69 BIBLIOGRAPHY The following papers contain descriptions of physical conditions and fea- tures and of some geologic observations in Porto Rico. A suggestion as to the range or character of the article in most cases follows the title. CLevE, P. T.: “Geology of the Northern West Indies,’ with maps. Kongl. Sevenska. Vetenskaps, Akad. Handlingar, vol. 9, No. 12. 1871. : “Outline of the Geology of the Northeastern West India Islands.” Ann. N. Y. Acad. Sci., vol. 2, pp. 185-192. 1883. Davis, Bric. GEN. GEORGE W.: Porto Rico—embracing reports of Brig. Gen. George W. Davis, arranged by topics. Government Printing Office, Wash- ington, 94 pp. 1900. Dinwippiz, W.: “Physical features of the Island” (Puerto Rico). MHarper’s Weekly, vol. 43, p. 248. Mar. 11, 1899. : “The great caves of Puerto Rico.” MHarper’s Weekly, vol. 48, p. 293. Mar. 25, 1899. DoMENECH, M. K.: “Mineral resources of Porto Rico.” Mines and Minerals, vol. 19, pp. 529-532. 1899. FALCONER, J. D.: “Evolution of the Antilles.” Scot. Geog. Mag., vol. 18, pp. 369-376, 1 pl. 1902. Discusses general geologic history of America, and especially that of Central America and West Indies. Guppy, R. J. L.: “Geological Connections of the Caribbean Region.” Trans. Canadian Institute, pp. 373-391. January, 1909. Discusses general re- gional relations and the probability of an Antillean continent. Fiske, Amos K.: “The West Indies.” G. P. Putnam’s Son, New York. 1899. Folded maps, plates. Physical characteristics of Porto Rican history ete., social and economic conditions. HAMILTON, S. HARBERT: “Notes on Some of the Ore Deposits of Porto Rico.” Eng. and Min. Journal, vol. 88, p. 518. September, 1909. Hitz, Ropert T.: “Cuba and Porto Rico with the Other Islands of the West Indies.’ New York. 1898. A book of over 400 pages containing a very large amount of information covering many matters in addition to the physical features and the geology. : “The Value of Porto Rico.” Forum, vol. 27, pp. 414-419. June, 1899. : “Porto Rico.” National Geographic Magazine, vol. 10, pp. 93-112. 1899. One of the best articles on the physical features of Porto Rico. : “The Geology and Physical Geography of Jamaica.” Bull. Museum Comp. Zool., vol. 34. 1899. s 7} 1 e eo a ae t 2 t 4 j “ be — + ‘ Ai) i Pr a a ee 7 he Z s ‘d i= 3 Since writing the above I have heard from Mr. Banks that he has seen it from Havana, Cuba. It is not included in the discussion at the end of this paper. 4 ANNALS NEW YORK ACADEMY OF SCIENCES ‘THERIDIIDE The members of this large family spin irregular webs and certain of them are very common about human habitations. Argyrodes is found throughout the tropics and subtropics. We took what Mr. Banks identified as A. /arvatus Keyserling in a narrow steep ravine near Banos San Vincente, Cuba, and an unidentified species, prob- ably the same, in a similar ravine near Merciditas, a few kilometers far- ther north. This species is now considered to be a synonym of cancellatus (Hentz) and the range is Connecticut to +Alabama, Venezuela and St. Vincent. A. nephile Taczanewski is known from southeastern United States, Peru, Brazil, +Guiana, Haiti and Bermuda. We found it on both Mona and Desecheo. A. trituberculatus Becker has been recorded from + Mississippi and Haiti. Rhomphea is found throughout most of the world’s tropics and sub- tropics. R. paradora (Taczanowski) was described from Guiana and has been reported from St. Vincent, but not elsewhere in the Antilles. Keyserling states that it is probably only a variety of fictiliwm (Hentz) which is the only known species in the United States and extends from ‘New England to Florida. Spintharus contains but two species, one confined to Brazil and one, fiavidus Hentz, which was described from Alabama and is now known from all the States to Peru and from St. Vincent. It is a yellow, red and black creature usually found on the under side of the leaves of small bushes and should be looked for in the Greater Antilles. Theridion is a large genus which is found throughout the world. Curiously enough, with the exception of certain cosmopolitan species, records from the Greater Antilles are almost lacking. We took antil- fanum Simon, hitherto only known from St. Vincent, by beating branches in a sea-grape thicket at Dorodo, near San Juan, Porto Rico, and also in a narrow ravine in the mountains north of Vinales, Cuba. TZ. fron- deum Hentz is found from Labrador (specimen in our collection) to Lower California and on St. Vincent. It will probably be found in the Greater Antilles. We took fuesslyi Simon, formerly known only from St. Vincent, by sweeping the tall grass and shrubs on the southwest coast, of Desecheo. Among the wide ranging tropical and subtropical species are rufipes Lucas, studiosum Hentz, tepidariorum C. Koch, and vituper- abile Petrunkevitch. We have the first two from the Bahamas but the only record for any of them from the Greater Antilles is that by Banks of studiosum Hentz from Haiti. It was described from South Carolina and Alabama and is known not only from “tropical and subtropical North LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 85 and South America” but also from Bermuda. Finally, volatile Weyser- ling is recorded from Florida, +Venezuela and (immature specimen) Culebra Island, Porto Rico.® Theridula is found in the Mediterranean region, Africa, Madagascar, tropical and eastern Asia, Japan, Philippines and in America from Lab- rador (epulenta in our collection) to Peru and Brazil. Simon gives the distribution of opulenta Hentz as the Mediterranean region, western Africa and the Transvaal, Madagascar, Ceylon, the Antilles and North America. It was described from Georgia. There are a number of synonyms which extend its distribution to Peru and complicate matters. Banks reported it, under the name of triangularis eyserling from Caya- mas, Cuba, and from Haiti. He also reported spherula (Hentz), prob- ably another synonym, from Haiti. He identified the specimens which we took in Cuba at Cabanas, Cerro de Cabras (near Pinar del Rio), Banos San Vincente, and Guane as triangularis, and others which we took at Cerro de Cabras as quadripunctata Keyserling. This latter is prob- ably also a synonym. Apparently there are no other records for the Greater Antilles. Latrodectus is found throughout most of the tropical and subtropical world, including Madagascar and Australia. The species tend to have bright red markings and “have been notorious in all ages and in all re- gions of the world where they occur on account of the reputed deadly nature of their bite’ (Cambridge). JL. mactans (Fabricius) extends out- side even the subtropics, being found from New York to Tierra del Fuego. Cambridge (Proc. Zodl. Soc. London, April, 1902, p. 253) points out that it is hardly separable from tredecim-guttatus, which is found in the Canaries, Madeira and the Mediterranean region. The only West Indian record is by Banks at Mayaguez, Porto Rico and Cockerell in Jamaica. We took it at Cabanas and Banos San Vincente in Cuba. L. geometricus C. Koch is found in Cape Verde Islands, Africa, Madagascar, India, Aus- tralia, Bermuda and tropical South America, the type locality being Co- lombia. The only West Indian records I have seen are Santiago de las Vegas and Havana, Cuba. Petrunkevitch gives the distribution of Teutana grossa (C. Koch) as “Mexico, Guatemala, Costa Rica, West Indies, Brazil, Uruguay, Chile, Argentina, Is. Juan Fernandez, (Europe, Africa).” I can find no au- thority for the West Indies and believe that he mistranslated Simon’s 6 Since writing the above I have heard from Mr. Banks that he has seen Theridion interruptum Banks and 7’. rufipes Lucas from Havana, Cuba. The former is known elsewhere only in Florida, but the latter is a cosmotropical species. They are not in cluded in the discussion at the end of this paper. 86 ANNALS NEW YORK ACADEMY OF SCIENCES “les iles de l’Atlantique,” although there is no reason for not expecting it in the West Indies. Lithyphantes is a cosmopolitan genus which has been recorded on the American mainland from Canada to Patagonia, but not from the An- tilles. We took septemmaculatus Kevserling in Cuba in a mangrove swamp hear Cabanas and im a flower garden at Banos San Vincente in the mountains. The localities given in the original description are stated as follows: “Herr Marx fing dieses Thier im Juli bei Denver in Colum- bia und im December bei Enterprise in Florida.” Doubtless Denver, Colorado, was intended. It is also recorded from Curacao. Mysmena is known from France, northern Africa, Cevlon, Philippmes and United States (two species, both in Florida, one of them being known also from the District of Columbia). We got several young speci- mens of the genus from the fallen leaves at the base of the cliffs at Banos San Vincente, near Vinales, Cuba. Theridionexus cavernicolus Petrunkevitch is the only species in its genus and is known only from the Peru Cave in Jamaica. Although he places this creature among the Theridiide, Petrunkevitch says: “Its gen- eral appearance, the long front legs and the globose abdomen, and most of all the presence of a well developed tarsal comb, speak for its close re- lation to the family Theridiide. On the other hand, the structure of the mandibles, the shape of the cephalothorax, and especially the presence of a tibial apophysis in the male palpus, are characters which are found only in the Argiopide. It is, therefore, impossible to place the genus Theri- dionexus with sufficient reason in either of these families; it forms a new, intermediate group.” From the standpoint of phylogeny, perhaps one might say it belongs to an old intermediate group, and if this be true its discovery,in a cave in the mountains of Jamaica has an added interest. LINYPHIID-® The distribution in America of this family is interesting. For the most part it is northern, but it has a number of representatives in south- ern South America and very few between. In addition to certain genera found only in Greenland or the extreme northern pait of the hemisphere and others found only in southern Patagonia or Tierra del Fuego, there are others such as Gonatium and Gongylidiellum which are found at both extremes but not between. This discontinuous distribution may be due to faulty taxonomy or there may be natural causes for it. It is, however, not unknown in other organisms or even in other groups of spiders. This family is sometimes considered to be a subfamily of Argiopide. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 87 Bathyphantes is known from Europe, Asia, Australia and New Zea- land, as well as from America. Of the forty American species, thirty-one are not recorded south of the United States, four are more southern but not known south of Guatemala, and four are from the extreme southern part of South America. The final one is the recently described sems- cincta Banks, collected by us in a flower garden at Bafios San Vincente, near Vinales, Cuba. Ceratinella is a palzearctic genus which has one species in this hemi- sphere. This species, brunnea Emerton, is found from Labrador (speci- men in our collection) to New York. We also have a specimen of the genus, species undetermined, from the pine-palmetto plains south of Pinar del Rio, Cuba. Ceratinopsis is an American genus with twenty-three species of which eleven are not recorded south of the United States (except, now, see anglicana) ; three from Mexico or Central America; and of the nine South American ones about half are confined to the southern part of that continent, three of them being known only from the region of Tierra del Fuego. We took anglicana (Hentz) by beating oak branches in a dense thicket at about 125 meters elevation on Cerro de Cabras, near Pinar del Rio, Cuba. We also took the genus, species undetermined, in a ravine at an elevation of about 300 meters near Banos San Vincente, Cuba. There are no other records for the genus in the West Indies. Linyphia is another genus which extends from the northern to the southern extremes of this hemisphere, in fact it is nearly world wide in its distribution, but it is rather better developed in the tropics than some of its relatives. The only West Indian species is coccinea Hentz, which is found in {Florida and Haiti. Microneta has a wide distribution, especially in temperate regions. In America there are twenty-one species confined to northern United States and Canada; one to Mexico; one (varia Simon) to St. Vincent; and one to Brazil. If the last two are correctly placed, the genus is likely to be found in the Greater Antilles. ARGIOPID This large family which includes the true orb-weavers is unsatisfactory material for a study of distribution because of the uncertain limits of some of the genera. Petrunkevitch and others have dodged the issue by putting eighteen of them in the Cohors Araneus, and I can only do like- wise, putting, however, the probable generic name in parentheses. One of the subfamilies (Linyphiine) into which Simon divides the Argio- pide has already been considered, treating it as a family. ANNALS NEW YORK ACADEMY OF SCIENCES CH oa) TETRAGNATHINZ Cyatholipus is a genus erected by Simon to contain two of his species from southern Africa and one, dentipes Simon, from Jamaica. It is so distinct that Simon considers it to be the representative of a special group of the subfamily. | Tetragnatha is an almost cosmopolitan genus. The known distribu- tion of antillana Simon is Mexico, Central America, Porto Rico (Lares), and St. Vincent. 7. elongata Waleckenzer may be fcund in the Greater Antilles as, although it is a species of Canada and United States, Walcke- ner has recorded it from Guadeloupe. It occurs near streams and usually hangs its web partly, at least, over the water. T’. laboriosa Hentz has been recorded from Alaska, much of the United States and Porto Rico (Utuado). Mr. Banks writes me that he has seen a specimen from Havana. TJ’. piscatortia Simon has been reported only from St. Vincent but we found it in Porto Rico at an altitude of about 500 meters on El Duque and also near Arecibo. TY. vicina Simon is recorded from +St. Vincent and Porto Rico (San Juan). Banks records the genus from Haiti on the basis of immature specimens and we have a number of such specimens from the edge of a mangrove swamp near Cabanas, Cuba. See also the discussion of Eugnatha. Eugnatha is probably not more than a subgenus of Tetragnatha but, following Banks and Petrunkevitch, it will be separately considered. There is one species apparently confined to Mexico and one to northern United States. HH. pallescens (F. Cambridge) is recorded from +New York, New Jersey, Florida, Texas, New Mexico, Mexico and Cuba (San- tiago de las Vegas and Havana). JF. gracilis Cambridge is recorded _ from Mexico, tGuatemala and Bayamon, Porto Rico. We found it at Naguaba, San Juan and Mayaguez in Porto Rico and at Banos San Vin- cente in Cuba. Meta, as construed by Simon and others, is a fairly large and almost cosmopolitan genus. As is pointed out below, Mr. Banks considers bigibbosa (Keyserling) to be a Leucauge and he expressed his further opinion that Meta does not occur in the tropics of America at least. Alcimosphenus has two species: bifurcatus Petrunkevitch from Jamaica; and licinus Simon. The latter is recorded from St. Vincent; Adjuntas, Porto Rico; Haiti and +Santo Domingo; Santiago de las Vegas, Cuba; and +Jamaica. We found it in Cuba at Cerro de Cabras and Banos San Vincente. Leucauge, more widely known as Argyroépetra, is a large genus which is generally distributed in the warmer parts of the world. JL. argyra LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 89 (Walckener) ranges from California and Florida to Brazil and is recorded also from Cuba (Havana), Haiti, Porto Rico (Aguadilla, Arroyo, Lares and El Yunque), Vieques, {Guadeloupe and St. Vincent. We took it at nearly all stations from Guane to Guantanamo in Cuba, at Hope Gardens and Montego Bay in Jamaica, on Mona, and at Arecibo and San Juan in Porto Rico. L. regnyi (Simon) has been recorded only from St. Vin- cent, although it is possible, as suggested to me by Mr. Banks, that Argy- roépetra bigibbosa Keyserling, which Petrunkevitch questionably put in Meta, is a synonym. If so, its range includes also Colombia, Porto Rico (Aguadilla and El Yunque) and Haiti. In that case bigibbosa is the prior name. We took it in Porto Rico in the foothills of El Duque near Naguabo and at Arecibo, on Mona and Desecheo, and very commonly throughout Cuba. JL. venusta (Walckener) is more generally known as Argyroépeira hortorwum (Hentz) Emerton. It is fairly common throughout the United States (Georgia being the type locality) and south to Colombia. Its recorded insular distribution is Bermuda, Bahamas, Cuba (Santiago de las Vegas) and St. Vincent. We did not take it in Cuba but did on Mona and in the mountains south of Arecibo, Porto Rico. As it is a very striking ‘“‘green and silver-white spider, tinged with golden, and sometimes with orange-yellow or copper-red spots” and usually rests in a conspicuous web, it is not likely to be overlooked. Dolichognatha is found in Ceylon, western Africa, and (the species given here) tropical and subtropical America. It is closely related to Diphya which has a somewhat similar distribution: South Africa, Mada- gascar, and Chile. D. tuberculata (WKeyserling) has been found in ¢Flor- ida, Mexico, Costa Rica, Venezuela and St. Vincent. It will probably be found in the Greater Antilles. NEPHILINA Nephila is found throughout much of the warmer parts of the world but seems to be rare in the West Indies, clavipes (Linnzus) being the only species recorded, although it has gone under a number of names, of which wtldert, wistariana and concolor, all by McCook, should be men- tioned. Its known range is from southern United States to Peru and Brazil, Bermuda, Bahamas, +Jamaica, Santo Domingo and Porto Rico (Aguadillo). We took an immature Nephila at Cabafias, Cuba, and as Mr. Banks has seen a specimen of clavipes from Havana our specimen doubtless belongs to this species. We have a number of adults from Mona. 90 ANNALS NEW YORK ACADEMY OF SCIENCES ARGIOPINA Argiope is found.in the warmer parts of the world, less frequently in the temperate. Three (aurantia and the two mentioned here) of the eight American species are wide ranging but the others are more re- stricted. A. argentata (Fabricius) is said to be found from southern United States to Patagonia, including all the islands. Among the latter are St. Vincent, Martinique, Culebra, Porto Rico (San Juan, Utuado and Aguadilla), Santo Domingo and Isle of Pines. We also have it from several stations in Jamaica. A. trifasciata (Forskal) is almost cos- mopolitan. Combining the records of Banks (fastuosa Olivier) and our captures, it is safe to say that it is found throughout Cuba and Banks had it from San Juan, Porto Rico, but I know of no other definite rec- ords from the Antilles. We have a number of undetermined specimens of the genus from Mona, Desecheo and Porto Rico (San Juan and Arecibo) which may be one, or both, of these species. Gea is known from West Africa, tropical Asia, East Indies, Polynesia, Australia and, by two species, America. One of these is reported only from Mexico. The other, heptagon (Hentz) has been found in tsouth- eastern United States as far north as the District of Columbia, in Guate- mala, Brazil and Cuba (Havana). We took it in Cuba at Pinar del Rio and Banos San Vincente. Cyrtophora is found throughout the tropical and subtropical world. It has three species in northern South America; one in California; and one, sellata Simon, in Santo Domingo. Cyclosa has a world-wide distribution. C. caroli (Hentz) is found from southern United States ({Alabama) to Venezuela; also in St. Vin- cent and at Havana, Cuba. C. oculata (Walckenzr) was described from BHurope (France and Italy) but Simon says it is probably of American origin and introduced to Europe where it is rare and localized. The American localities given by him are Venezuela and “Antilles.” It would be interesting to know which of the Antilles are concerned and how abundant and widespread the species is in this hemisphere. ‘The re- corded distribution of C. walckeneri (Cambridge) is California, Mexico to +Colombia, Brazil, Cuba (Santiago de las Vegas) and Haiti. We took it in Cuba also at Cerro de Cabras near Pinar del Rio. Edricus is an American genus with a range from Mexico to Peru and Brazil. We took crassicauda (Keyserling) in the mountains about mid- way between Arecibo and Utuado, Porto Rico. ‘This species is found on the mainland from Mexico to +Colombia. Mangora has a wide distribution but has not been reported from Africa, Madagascar, the Pacific Islands or Australia. Neither has it LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 91. previously been reported from the West Indies although it is known to extend from northern United States to Brazil. We took a common United States species, placida (Hentz), in the Cerro de Cabras near Pinar del Rio, Cuba. This is not a species which would be likely to be carried by commerce, and although the United States Army operated to some extent in these hills it is not probable that this is a human intro- duction. If it were we would have expected to find the species nearer the regular lines of traffic. Alabama is the type locality. Acacesia foliata (Hentz) is the only species in the genus as construed by Petrunkevitch. It is recorded from New England to Panama and from Haiti. It was described from Alabama. Eustala is a fairly large American genus, chiefly found in Central America but ranging throughout most of the mainland south of Canada. FE. anastera (Walckener) has many synonyms, the most common of which is Hpeira prompta. It was described from Georgia and its distribution includes most of the United States and southward to Costa Rica and the Galapagos; Cuba (Cayamas and Havana according to Banks and Guane, Pinar del Rio, Banos San Vincente and Santiago in our collection), Haiti, Porto Rico (2800 feet altitude on El Yunque according to Banks and Arecibo in our collection) and St. Vincent. The variety, or separate species, conchlea (McCook) is known from +California, +Florida and Utuado, Porto Rico. The recorded distribution of F. fuscovittata (Key- serling) is Mexico to Paraguay, Guatemala being the type locality, Cuba, Porto Rico and St. Vincent. We found it in nearly all parts of Cuba and Porto Rico which we visited but not on Mona or Desecheo. Drexelia is probably a synonym of Larinia, a genus found throughout most of the subtropical, and, more rarely, tropical world. As limited here (following McCook, Cambridge and Petrunkevitch) there are two species in America: one in Mexico, and one, directa (Hentz), recorded from +southeastern United States to Panama and at Havana, Cuba. We took it in several places on the plain south of Pinar del Rio, Cuba. Cohors Araneus. As was stated above, this group includes a number of genera whose limits have not been clearly defined. A. (Neoscona) arabesca (Walckener) is found on the mainland from . Labrador to Mexico and also on Curacao. We took it in western Cuba at Pinar del Rio and Guane by sweeping in meadow land. It was de- scribed from the Carolinas. A. (Verrucosa) arenatus (Walckener) was described from Georgia and is known from New Jersey west to California and south to Panama. McCook says: “I have specimens collected by the late Mr. W. H. Gabh from San Domingo varying in but slight particulars from those above described.” 92 ANNALS NEW YORK ACADEMY OF SCIENCES A. (Eriophora) balaustinus McCook is recorded from +Florida to Lower California and Mexico; San Juan, Porto Rico; Haiti and Santo Domingo; Cayamas and Havana, Cuba; Jamaica; and Swan Island. The latter island, lying between Cuba or Jamaica and Central America, is a very interesting locahty. It is to be regretted that we know so little of its fauna. We took this species on Mona and in a rotten banana trunk near Arecibo, Porto Rico. A. (Neoscona) benjaminus (Walckener). lJLabrador (in our collec- tion) to Guiana and in Martinique. It is probably in the Greater Antilles. A. (Epeira) bispinosus (Keyserling) is recorded from +Calitornia, Arizona, Panama and Haiti. A, (Singa) crewii (Banks). +aiti. A, (Singa) cubana (Banks). The type locality is Havana, Cuba. We took it on the same island at Pinar del Rio and Banos San Vincente by sweeping the grass at the edge of water. A. (Neoscona?) granadensis (Keyserling) is recorded from +Colombia and the Luquillo, Porto Rico. In giving the latter record, Banks remarks that it is close to trivittata Kevserling which is here considered to be a synonym of arabesca. ! A. (Epeira) gundlachi (Banks) was described from a specimen taken by us on the sandy plain about 12 kilometers south of Pinar del Rio, Cuba. A. (Epeira) incertus (Cambridge), described from Costa Rica, has hitherto been known only from Mexico and Central America. We got it in Cuba at Cabanas and Esperanzia on the edge of mangrove swamps, in the mountains north of Vinales on oaks, and south of Pinar del Rio on the palmettoes of the dry sandy plains. A species so catholic in its eco- logical tastes is doubtless more widely distributed than these records show. A. (Metepeira) labyrintheus (Hentz) is said to occur “from Labra- dor to Patagonia, including all the islands.” The only definite Antillean records I have seen are Barbados; St. Vincent; Culebra and Vieques; Aguadilla, Porto Rico; and Havana, Cuba. We took it at Mayaguez, Porto Rico, and on Desecheo. Its type locality is North Carolina. A. (Verrucosa) mexicanus (Lucas) of which wndecimvariolata Cam- bridge is considered to be a synonym is known from Panama, Costa Rica, +Guatemala, Mexico and Santo Domingo. A. (Neoscona) nauticus (1. Koch) was described from Africa. It is said to be found throughout the world’s tropics and also in New Hamp- shire and Tennessee. Its only Antillean records are St. Vincent and Haiti. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 93 A. (Neoscona) neotheis Petrunkevitch is a name proposed for theisii of Keyserling and McCook but not theis of Walckener which is a Poly- nesian species. If this synonymy be followed, the known distribution is California, Mexico, Guatemala and probably the followimg Antillean lo- ealities,—Culebra; El Yunque, Bayamon, Aguadilla and Mayaguez, Porto Rico; and Haiti. In givimg the Porto Rican records Banks says that “[thetsw| is smaller and more slender than H. oaxacensis Keyser- ling, and I think different, although small specimens of the latter look much like large specimens of the former.” See the next species. A. (Neoscona) oaxacensis ( eyserling) is, as is indicated above, con- fused with neotheis. Petrunkevitch gives its distribution as Pacific Coast of United States, +Mexico, Panama and St. Vincent. Banks recorded it from Santiago de las Vegas, Herradura and Havana in Cuba but made theist Keyserling a synonym. Specimens taken by us at Cabanas and Pinar del Rio in Cuba, on Mona, and at Mayaguez, Arecibo, Manati, San Juan and Naguabo in Porto Rico were identified by Mr. Banks as this species. A. (Epeira) pegnia (Walckener) is recorded from Colombia, Costa Rica and most of the United States. Mr. Banks has written me that he has seen Hpeira globosa Weyserling, here considered to be a synonym, from Havana, Cuba. A. perplexus (Walckener) was described as an FP peira in 1842 from Brazil and Santo Domingo but has, apparently, not been recognized since. A. (Epeira) pratensis (Hentz) is recorded from most of the United States but not elsewhere. We took it near Banos San Vincente, Cuba, and Mr. Banks informs me that he has seen it from Havana. A. (Epeira) sericatus Clerck apparently has a number of synonyms. Among them is vulgaris Hentz, the name which Mr. Banks gave to our specimens from Cabanas, Pinar del Rio and Guane, Cuba, as well as to those he had from Santiago de las Vegas and Havana in the same island. In all cases they were about houses and may have been introduced. The only other American records are from the continent north of Mexico. It is also found in + Europe. A. (Marxia) stellatus Walckener was described from southeastern United States and is found from Labrador (specimen in our collection) to Guatemala. We took it on the plain ten kilometers south of Pinar del Rio, Cuba, at the edge of a swampy area. A. (Wagneriana) tauricornis (Cambridge) was described from a number of localities in Guatemala and Panama. It is recorded also from Colombia, Mexico, Louisiana, Alabama, Florida, Cuba (Santiago de las Vegas) and Haiti. 94. ANNALS NEW YORK ACADEMY OF SCIENCES A. (Wagneriana) undecimtuberculatus (Keyserling) is known from tColombia, Panama, Guatemala, Mexico, Florida, Cuba (Santiago de las Vegas) and Haiti. A. (Eriophora) variolatus (Cambridge) is found in southern and western United States, Mexico, +Guatemala, Venezuela and St. Vincent. It is probably in the Greater Antilles also. A. (Metazygia) wittfelde (McCook) is known from +Florida, Mexico, the Bahamas and Havana, Cuba. We took it at Mayaguez, Porto Rico, in a hotel. ; Gasteracantha is widely distributed in the tropical and subtropical regions of the world. G. cancriformis (Linneus) is found from North Carolina and California to Paraguay; the Bahamas; El] Guama, Santiago de las Vegas and Havana, Cuba (according to Banks under the name of hexacantha Fabricius) ; Haiti and Jamaica. We have it from the Ba- hamas, Jamaica, Mona and Banos San Vincente in Cuba. G. hilaris Thorell is recorded from ¢St. Bartholomy, Porto Rico (Aguadillo and Adjuntas), and Haiti. In Porto Rico we found the under side of leaves of young coco palms at San Juan and also in the mountains south of Arecibo festooned with its webs. Banks has referred G. canestrinti Cam- bridge to this species, thus extending its range to Antigua and Dominica, but Petrunkevitch does not follow him in this. G. sexserrata (Walek- ener) is recorded from +Cayenne, Haiti and Porto Rico (Bayamon). I took it several years ago on Key Largo, Florida, so it is probably in Cuba also. G. tetracantha (Linneus) is recorded from California (if pallida Koch be a synonym), Haiti, Culebra, St. Thomas, St. Vincent, Guade- loupe and three of the Grenadines. We can now add to this list. Porto Rico (near Arecibo) and Desecheo. Micrathena is the Acrosoma of authors. As now constituted it con- tains more than 125 species, all American, and the great majority trop- ical. M. armata (Olivier) is known only from + Hispaniola and Jamaica. M. cubana (Banks) is recorded only from +San Diego de los Baftos and possibly Havana, Cuba. We took it in Cuba at Banos San Vincente. M. flavomaculata (Keyserling) is known only from Haiti; and forcipata (Thorell) only from Cuba. M. horrida (Taczanowski) is recorded from Mexico, Peru, +Guiana, Brazil and Jamaica; obtuspina (Keyserling) from + Mexico to Brazil and also from Porto Rico. Finally, rufopunctata (Butler) and sloanei (Walckenzr) are known only from Jamaica. It is decidedly curious that there are no records of the genus in the Lesser Antilles. Glyptocranium is strictly an American genus. The related Old World genera are found in Australia, Africa and southern Asia. It has two LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 95 species which are found only north of Mexico (one ranging even to Alaska), two in Mexico and one, gastercanthoides (Nicolet), in Brazil, tChile and Jamaica. Epecthinula minutissima Simon is the only species of its genus and is known only from Jamaica. The genus is related to Hpecthina, which lias but a single species, found only in Venezuela. According to Simon, there are three other genera belonging to the same group of the Argiop- ine. One of these is found only in northern South America; one in South Africa and western Australia; and one in Algeria, New Caledonia and northern South America. | MIMETID This is a small, for the most part tropical, family the members of which are usually found near or on the ground and make no definite web. None have heretofore been recorded from the West Indies. Mimetus is found in the Mediterranean region, Africa, India, New Zealand and America. In the latter hemisphere there are nine species, of which two are recorded only from South America, five from Mexico and Central America, one from North Carolina. The remaining one, mterfector Hentz, ranges over much of United States and we took it on the sandy plain south of Pinar del Rio, Cuba. The genus occurs also in Porto Rico, for we took’an immature specimen of it near Arecibo. THOMISID® These are called crab-spiders, not only on account of their shape but also because they usually walk sidewise or backward. They spin no snares but catch their prey either by pursuit or by Jying in wait for it, being much favored with concealing colors. It is a large family the American members being chiefly found in or near the tropics although it ranges from Greenland to Patagonia. MISUMENINZ Misumenops is a split, possibly unwarranted, from the cosmopolitan Misumena, to include a number of American species. Jf. americanus (Keyserling) is recorded from United States, Guatemala and St. Vin- cent. M. asperatus (Hentz) ranges from Canada to Costa Rica and is known also from Cuba, Haiti, Porto Rico (Bayamon and El Yunque) and St. Vincent. We found it very common throughout Cuba and took it also on Mona and at Mayaguez and San Juan, Porto Rico. J. bellulus (Banks) is recorded only from Florida but Mr. Banks has written me 96 ANNALS NEW YORK ACADEMY OF SCIENCES that he has it from Havana, Cuba. J/. celer (Hentz) probably includes M. spinosa Keyserling. Mr. Banks has recorded spinosa from Santiago de las Vegas and Cayamas, Cuba, and has named the specimens which we took at nearly all of our stopping places in that island from-Guane to Guantanamo spinosa, while he named the specimens which we took at Mayaguez, Arecibo, Manati and Naguabo in Porto Rico celer. On the mainland, celer is found from Massachusetts to Mexico. ‘The known dis- tribution of oblongus (Keyserling) is from Massachusetts to Illinois and south to Georgia and New Mexico; also in Cuba (Santiago de las Vegas). M. viridans (Banks) is recorded only from Florida but Mr. Banks has it from Havana, Cuba. See also the next genus. Misumessus echinatus Banks was described from material beaten from oak trees on Cerro de Cabras near Pinar del Rio, Cuba. According to the system followed here, it should probably be put under Misumenops. STEPHANOPSINA Isaloides contains but two species: one from Mexico and one, tous- saintu Banks, from +Haiti and Cuba. It is closely related to and con- sidered by Simon to be not more than a sub-genus of Hedana. He would also include the South American genera Diea and Erissus. This adds interest to the distribution data, one section of the genus being found only in Hispaniola and South America while the other section is known only from Ceylon, southwestern Asia, Philippines, Malasia, the Tonga Islands, Australia and New Zealand. Stephanopsis is confined to Madagascar, Malasia, Melanesia, Poly- nesia, Australia and America. In America there is one species in Pan- ama, eight in South America (chiefly Chile) and one in Tortola of the Virgin Islands. Banks has recorded the genus from Santiago de las Vegas, Cuba. He states that he had two species, both represented only by Immature specimens, and suggests that one of them is a young pentagona \Keyserling. This species is now considered to belong to Onocolus. ‘The species is known from Panama, Peru and Brazil. and as Onoculus is solely a South American genus the interest in the distribu- tion of Stephanopsis is not lessened. Tobias is an American genus, four species being confined to northern South America and one, rugosus (Taczanowski), being recorded from Brazil, +Guiana, Peru, Central America and Haiti. PHILODROMINZA Tibellus is widely distributed in the temperate regions of the world, less so in the warmer regions. In America there is one species confined LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 97 ‘ to each of Mexico, Guiana and Paraguay while the fourth species, ob- longus (Walckener), occurs in +Hurope, Asia, Alaska and most of United States. We took the latter species on the dry plain nine kilometers south of Pinar del Rio, Cuba. CLUBIONID-® The Clubionids are frequently confused with the Drassids and their habits are much the same. They make nests in rolled leaves, under stones, or in rubbish. ‘The alphabetical arrangement of genera used by Petrunkevitch is especially confusing in this family as there are very distinct divisions which may be of family rank. The arrangement of Simon will be followed here. SELENOPINZ Selenops is the only genus. It is found throughout most of the world’s tropics. They are flat creatures which crawl under bark or stones whence they dart out for their prey. S. atssus Walckener is known from Bahamas, Tortugas, Cuba (Cayamas and Santiago de las Vegas) and +Martinique. We found it hiding back of the boards of a house at Cabanas, Cuba. Macleay in describing S. celer (Macleay) said that it is common in Cuba. We took it north of Vinales, Cuba, on banana trees. It is recorded only from this island and Buen Ayre, off the coast of Venezuela. S. insularis Keyserling is known from +Porto Rico (San Juan, at least) and Haiti. We took it on Desecheo in a rotten log and also under fallen leaves in a sea-grape thicket. SPARASSINA Heteropoda is also found throughout most of the world’s tropics. There are only four American species: three confined to South America and the cosmo-tropical venatoria (Linneus). The known West Indian distribution is Jamaica, Cuba (apparently throughout), Haiti, Porto Rico (San Juan) and St. Lucia. It is frequently brought north in fruit. Olios is likewise a cosmo-tropical genus. O. antiguensis (Keyserling) is known from Haiti, Porto Rico (Utuado), Culebra and +Antigua. 0. bicolor Banks was described from specimens which we took on Desecheo, Mona, and at San Turce near San Juan, Porto Rico. O. maculatus (Blackwell) is reported from Brazil and the “Antilles.” Pseudosparianthis is an American genus with one species in Mexico, two in Brazil, one in St. Vincent, and one, cubana Banks, in Cuba (+Havana and +Santiago de las Vegas). 98 ANNALS NEW YORK ACADEMY OF SCIENCES CLUBIONINZA Anyphena in America is largely northern but it does extend even to Patagonia. It is found also in Japan, the mountains of India and the western Mediterranean region. We took immature specimens of it near Cabanas, Pinar del Rio and Guantanamo in Cuba. A. perpusilla Banks is known only from Santiago de las Vegas, Cuba. A. striata Becker was described from Mississippi and is elsewhere known only by Banks’s record, under the genus Aysha, from Santiago de las Vegas, Cuba. He also records under the same generic name A. velox Becker from Havana, Cuba. It is known also from + Mississippi, Florida and the Bahamas. A. graci- lipes Banks is known only from Haiti. Aysha is a small tropical American genus closely related to the pre- ceding one. Nine species are found in the region from Central America to Brazil; one is confined to the Galapagos Islands; two, ferox Simon and ravida Simon, are recorded only from Santo Domingo; and, finally, tenuis (lL. Koch) is known from Cuba, + Haiti, Santo Domingo, Porto Rico (San Juan), Culebra and St. Vincent. We took tenwis on Desecheo and obtained immature specimens of the genus on Mona. Chiracanthium inclusum (Hentz) was reported by Banks from San- tiago de las Vegas, Cuba and Aguadilla, Porto Rico. We took it at Guane and at Guantanamo, Cuba, but not between these two places. We also took it in Porto Rico near Quebradillas, in the seed pods of Crotolaria retusca which had been eaten out by Lepidopterous (Utetheisa) larve, and near Dorodo. It has been found in St. Vincent, Haiti, throughout’ much of tUnited States and in Mexico. The genus is nearly world-wide im its range. Clubiona-is another cosmopolitan genus but it is more abundant in temperate than in tropical regions. In America it extends from Lab- rador to Patagonia. Its best South American development is in Chile. (’. maritima I. Koch is reported from Santiago de las Vegas, Cuba, and +St. Thomas. C. pallens Hentz occurs on the Atlantic Coast from Massa- chusetts to tAlabama, and we found it in a weedy meadow on Cerro de Cabras, Pinar del Rio, Cuba. Eutichurus is reported only from Central and South America except for insulanus Banks, which is in Cuba, Haiti and +Bermuda. Oxysoma has ten species in South America. As they are chiefly in the southern and western part, it is interesting that the eleventh (cubana Banks, omitted in Petrunkevitch’s catalogue) should be found in Cuba. It was described from one male from Havana and has not been recorded elsewhere. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 99 Wulfila is an American offshoot of Anyphena. It is known only from Mexico, Central America, Bahamas and West Indies. There is one species recorded only from St. Vincent. W. immaculata Banks was de- scribed from specimens collected by us at the edge of a mangrove swamp near Cabanas, Cuba; on the dry coastal plain of Mona; and at an alti- tude of about 500 meters on El Duque, near Naguabo, Porto Rico. W. pretiosa Banks was described from a specimen which we took in the Rio Seco woods on the San Carlos Estate near Guantanamo, Cuba. W. parvula Banks is confined to Haiti and tenwissima Simon to Jamaica. CTENINA Ctenus is a cosmo-tropical genus. America has a large number of species but apparently none of them live north of southern United States and the Bahamas. St. Vincent and St. Lucia each have a peculiar species; haitiensis Strand is known only from Haiti and malvernensis Petrunkevitch only from Jamaica. We have the latter from Montego Bay, Jamaica, and unidentified specimens of the genus from Cuba (Banos San Vincente, Cerro de Cabras and Guantanamo). Banks recorded an immature Ctenus which “looks very much like Ct. hibernalis Hentz” from Santiago de las Vegas, Cuba. This species is known from Alabama, New Mexico, Mexico and the island of Buen Ayre. He also recorded a female Clenus (Microctenus) which appears to be new from a cave near Pueblo Viejo, Porto Rico. Cupiennius is a South and Central American genus except for one species which comes as far north as Florida and cube Strand which is known only from Cuba. It is interesting that, with the exception of widely distributed Ctenus, enly one genus of this division (Ctenez) of the Ctenine is found outside of the middle portion of America. That is Uliodon, which is known only from Madagascar, New Zealand and Australia. | LIOCRANING Syrisca is found in Africa and America. There is one species in each cf Colorado, Utah and Texas; two in South America, and two in the Greater Antilles. S. insularis (Lucas) is known only from Cuba and keyserlingt Simon (= Teminius insularis, Keyserling) only from + Haiti and Santiago de las Vegas, Cuba. MICARIINZ Castaneira occurs in the western Mediterranean region, Africa, central Asia and America. It appears to be better developed in United States 100 ANNALS NEW YORK ACADEMY OF SCIENCES and Mexico than it is farther south. The only West Indian record I have seen is a species confined to St. Vincent. We took a common United States species, descripta (Hentz). in a valley near Banos San Vincente, Cuba. CORINNINZA Corinna is known from Africa, tropical Asia, Malasia and America. It is well developed in this hemisphere but, although there are eight spe- cies in St. Vincent which are not known elsewhere, only two species have been recorded from the remaining Antilles. C. gracilipes (Keyserling) is known from Havana, Cuba, and from +Haiti. C. humulis (Keyserling) has been recorded from Haiti, Porto Rico (Luquillo and Hacienda Es- paranza) and +St. Kitts; Mr. Banks has written me that he has seen a specimen from Havana, Cuba; and we have a specimen from Mayaguez, Porto Rico, which is probably this species. Trachelas occurs in the Mediterranean region, Africa, Madagascar, India, Malasia and America. In this hemisphere it appears to be best developed in Central America, but bicolor Keyserling is the only species recorded from the Greater Antilles. It is known only from Cayamas and Santiago de las Vegas, Cuba, and from + Haiti. AGELENID © These are popularly called funnel-web spiders because many of the species spin flat webs which have a funnel-like passageway to their re- treat. In America the family is best developed in the north and prob- ably next best developed in the extreme south, being rather weak in the tropics except in the more mountainous regions. Hahnia occurs in Europe and the Mediterranean region, the moun- tains of India, Sumatra, Japan and America. There are seven species not found south of the District of Columbia, one of them being known only from Greenland ; two species are in Patagonia and Tierra del Fuego; and one, ernstt Simon, recorded only from St. Vincent, but taken by us in a rotten banana stump at about 500 meters elevation on El Duque, near Naguabo, Porto Rico. Tegenaria is found throughout most of the cooler and, more rarely, in the warmer portions of the world. T. domestica (Clerk), better known as derhami (Scopoli), is.said to inhabit the dwellings of man in all re- gions of the world, but I know of no definite record from the Antilles. T. insularia Walckener is known only from Cuba and has not been re- corded since its description. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 101 PISAURID The females of these, like those of the Lycosids, carry their egg sacs about with them. ‘Some species also build a “nursery” for the newly hatched young, but construct no snare. Thanatidius has one species in the Amazon region. The only other one known is dubius (Hentz), which has been reported from +North Carolina, Alabama, Florida and Cuba (Havana). Thaumasia, as now constituted, is found only in the warmer parts of America. TJ. marginella (C. Koch) Simon is usually placed in Dolo- medes and is recorded from Colombia, Brazil, Jamaica, Haiti, Vieques and possibly Porto Rico. We found it at Cabanas, Pinar del Rio and Banos San Vincente, Cuba. Mr. Banks has written me that he has seen it from Havana, Cuba. As he considers this species to be a Dolomedes, it may be that the immature specimens taken by us at Pinar del Rio and identified by him as Dolomedes belong to this species. As now construed, there are no records for Dolomedes in the Antilles. LycosI1pz& The relatively poor development in the Antilles of this fairly large family of “ground spiders” seems to accord with the distribution of Coleoptera (see Leng and Mutchler, 1914), among which the ground forms are more poorly represented than the arboreal ones. Bates, in his “Naturalist on the River Amazons,” says: “It is vain to look for the Geodephaga, or carnivorous beetles, under stones, or anywhere, indeed, in open, sunny, places. The terrestrial forms of this interesting family, which abound in England and temperate countries generally, are scarce in the neighborhood of Para—in fact, I met with only four or five spe- cies; on the other hand, the purely arboreal kinds were rather numerous. The contrary of this happens in northern latitudes, where the great ma- jority of the species and genera are exclusively terrestrial. . . . The remarkable scarcity of ground beetles is doubtless attributable to the number of ants and Termites which people every inch of surface in all shady places and which would most likely destroy the larve of Coleop- tera. These active creatures have the same functions as Coleoptera, and thus render their existence unnecessary. The large proportion of climb- ing forms of carnivorous beetles is an interesting fact, because it affords another instance of the arboreal character which animal forms tend to assume in equinoctial America, a circumstance which points to the slow adaptation of the Fauna to a forest-clad country, throughout an immense lapse of geological time.” The last suggestion, in a modified form, seems 102 ANNALS NEW YORK ACADEMY OF SCIENCES to be important. It may be that the fauna of “equinoctial America” is, to a large extent, a relic of past ages; that it arose in a forest-clad earth and has been driven before the more successful ground-inhabiting forms, not that it has been slowly developing from ground-inhabiting forms in the region in which it is now found. Lycosa is a large genus of world-wide distribution. Several attempts have been made to split it up, with varying success. One of these puts cinerea (Fabricius) in the genus Arctosa. This species has been found in Europe, throughout United States, in Mexico and at El Guama, Cuba. L. atlantica Marx was known only from Bermuda, but Mr. Banks writes that he has seen it from Havana, Cuba. JL. aussereri (Keyserling) is known from ¢Colombia, Vieques and Culebra; L. badia (Keyserling) from Central America and Cuba; L. fusca (Keyserling) from Central Amer- ica, Cuba (Havana, according to a letter from Mr. Banks) and Porto Rico (San Juan); Ll. punctulata Hentz, from United States east of the Rockies and, according to a letter from Mr. Banks, Havana, Cuba. L. insularis Lucas is known only from Cuba. We took it there at Guane, Pinar del Rio, Banos San Vincente, Cabanas and Guantanamo. Banks, who now considers it to be a Pardosa, recorded it from Santiago de las Vegas. Pardosa is likewise of world-wide distribution. In America it is best developed in the north. There are a few species in South America, but unless insularis be included the only West Indian species is portoricensis Banks, which was described from San Juan, Porto Rico. We took it ina inarsh at San Turce near its type locality. These two genera, while much confused in the literature, are placed in different divisions of the family by Simon. Both are too generally dis- tributed to be of a great deal of interest in a study of distribution. OxYOPIDZ The members of this family run about the vegetation with great agility in chase of their prey. There are only eight genera, of which six occur in America. Of the remaining two, one is confined to Madagascar and one io India and Malasia. Two of the six American genera have not been reported from the Antilles; one of them has but two species and is con- fined to Brazil, and one has but one species, being confined to Cayenne, as far as is known. Hamataliva has a wide tropical and subtropical range but in America there are only four species known: two in Brazil, one in Mexico and grisea Keyserling, which is reported from Lower California, southern: United States. Cuba (Santiago de las Vegas) and Haiti. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 103 Oxyopeidon is found in tropical east Africa, India, Indo-China and America. With us it seems to be confined to Mexico and Central America (sIx species) except for rana Simon, which has been known only from St. Vincent. We found this species fairly abundant on low vegetation on Mona. Oxyopes is a nearly cosmo-tropical genus which extends into more temperate regions. In America it is best developed in Mexico and Cen- tral America, where there are numerous species. O. pallidus (C. Koch) was described from the West Indies, but the only definite locality record seems to be in Walckener’s description of a male from Cuba. O. salticus Hentz is known from New York to +Alabama, Kansas, California and south to Bolivia, also in Bermuda, all the Greater Antilles and St. Vin- cent. It seems to be common throughout Cuba. We found it on Mona and near San Juan, Porto Rico. In Porto Rico it is also reported from El Yunque and Culebra. Peucetia is also found throughout most of the tropical and subtropical regions of the world. P. poeyi (Lucas) is known only from Cuba, but viridans (Hentz) is distributed in America from +North Carolina to Cali- fornia and south to Costa Rica. It is recorded from Cuba (Santiago de las Vegas and Cayamas), Haiti and Jamaica. We have it from Kings- ton, Jamaica; Guane, Banos San Vincente, Oriente and Guantanamo in Cuba; and Mayaguez, Porto Rico. ‘It has, apparently, not been found in the Lesser Antilles. SALTICID-® The “jumping spiders” make no snare but run about freely on the ground and on the vegetation in pursuit of their prev. The family is a large one and contains many species still to be described. while the taxonomy of the known forms is in a far from satisfactory shape. Our own West Indian collection was in the hands of Mr. Peckham and only partly worked up when he died. The following records, therefore, do not entirely cover the material which we have on hand. Agobardus anormalis Kevyserling is the only species in its genus. The type specimens are labeled U. S., but Petrunkevitch, following Peckham and Banks, states that it is probably from the West Indies. Bythocrotus cephalotes (Simon) is not only the only species of the genus but the genus is the only one in Bythocroteew, one of Simon’s divisions of the Salticide. It is not known outside of Haiti. Compsodecta is an American genus with one species in Guatemala and two, namely, albopalpis (Peckham) and grisea (Peckham), in Jamaica. The genus belongs to Simon’s group Pensacoleew which con- 104 ANNALS NEW YORK ACADEMY OF SCIENCES tains but two other genera, one of which is best represented in Brazil and Ecuador but extends to Mexico and the other is confined to tropical western Africa. ; Corythalia is a fairly large tropical American genus. ‘Two species, major (Simon) and sellata (Simon), are mentioned in Simon’s “Histoire Naturelle des Araignees,” II, pages 655 and 649, in such a way that Petrunkevitch was led to credit them to the West Indies. and, while this seems to me questionable, I can find no more definite reference. C. metallica (Peckham) is reported only from St. Vincent. Although it does not properly come within the scope of this paper, it is worth men- tioning that we have it from Dominica. C. elegantissima (Simon) is known only from Santo Domingo, and locuples (Simon) from both parts of Hispaniola. Stoidis is also an American genus. It contains but two species at present but specimens taken by us in Cuba were marked by Mr. Peckham as new species. 8S. aurata (Hentz) is known from +South Carolina and Florida. Mr. Banks has informed me that he has seen it from Havana, Cuba. S. pygmea (Peckham) is reported only from St. Vincent but we took it in an epiphyte, Ti//andsia utriculata, on Mona. These two genera, Corythalia and Stoidis, are the only American rep- resentatives of Simon’s Zenodorew. The other four genera are from Australia and the Pacific Islands. The Cohors Dendryphantes of Petrunkevitch includes, in addition to some unplaced species which should probably be put in the genus Dendryphantes. the following genera: Metaphidippus, Paraphidippus, Parnenus and Phidippus. Although this cohors is in large part the genus Dendrypiantes Koch as construed by Simon it is not very satis- factory for a study of distribution. The following list is arranged ac- cording to Petrunkevitch. Dendryphantes armatus Banks (omitted by Petrunkeyitch) was de- scribed from Santiago de las Vegas, Cuba, and has not been reported since. D. (Phidippus) audax (Hentz) is known from Canada, throughout the United States and from Santiago de las Vegas, Cuba. _ D. (Paraphidippus) aurantius (Iucas} is found from southern United States to Costa Rica and in Santo Domingo. D. (Metaphidippus) capitatus (Hentz). Canada to Mexico and Santiago de las Vegas, Cuba (in Banks’s list under name of octavus Heniz). Dendryphantes mendicus (C. Koch) is credited to “West Indies” but no definite data is given. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 105 D. (Phidippus) miniatus (Peckham). The known range is Virginia to Texas and at Santiago de las Vegas, Cuba. D. (Phidippus) octopunctatus (Peckham) is reported only from Missouri but I took it in Dominica, the identification being by Peckham. It is, therefore, probably in the Greater Antilles. D. (Metaphidippus) proximus Peckham was described from Cuba. Banks lists it from Santiago de las Vegas and we took it at Santiago de Cuba, Cristo, Zaza del Media, Cabanas, Cerro de Cabras and Banos San Vincente, Cuba. We have it also from Kingston, Jamaica. D. (Metaphidippus) prudens Peckham is known only from Kingston, + Jamaica. D. (Phidippus) regius (C. Koch) is known only from Cuba. Banks reports it at Santiago de las Vegas and we took it at Guane, a number of places in the vicinity of Pinar del Rio, north of Vinales (Banos San Vincente and Merciditas), Cabanas and Zaza del Media. All these localities are in central and western Cuba. D. (Metaphidippus) taylori Peckham is known only from Jamaica. Evophrys is found in Europe, Africa, Japan and America. There are about fifty American species, the genus ranging from New Hampshire to Patagonia. One species is found in St. Vincent, another in St. Thomas; it may, therefore, be found in the Greater Antilles but there do not seem to be any records as yet. The only other genus of Evo- phrydez is confined to Central America and northern South America. Eustiromastix has seven species; three are confined to Brazil, one to Colombia, two to St. Vincent and haytiensis Banks to Haiti. Hasarius adansoni (Adouin) is found throughout much of the world’s tropics but I know of no West Indian records. It is the only American representative of the genus with the possible exception of bellicosus Peck- ham from Guatemala. Hyctia is recorded only from Europe and United States. One of the three American species, pike Peckham, ranges from New England to New Mexico. We took it in Cuba at Cabanas, on the plains south of Pinar del Rio, and at Guane. Mr. Banks has written me that he has seen it from Havana. Icius is given by Simon as being found in Europe, Africa and Asia but he does not give America. We have thirteen species, all found north of Mexico except one which is confined to Uruguay, one to Florida and the Bahamas, and seperatus Banks which is known only from Haiti. Lyssomanes is a large American genus ranging from southern United States to Brazil. The. Lyssomanee includes seven genera, of which two are confined to America (the other one being found from Central America 106 ANNALS NEW YORK ACADEMY OF SCIENCES to Guiana), two to Madagascar, one to India and Ceylon, one to the Philippines, and one is found in western Africa, Madagascar, the Sey- chelles, Cevlon, Burma and Indo-China. The group is evidently an an- cient one. We have an undetermined specimen of Lyssomanes from Porto Rico. JL. antillanus Peckham is known from both parts of Hispaniola and we have it from Jamaica (Lapland). JL. viridis (Walckener), the most northern species of the genus, is known from tsouthern United States, Central America, Haiti and (by letter from Mr. Banks) Havana, Cuba. Marpissa occurs in Europe, Asia including Japan, and America. On the occidental mainland it ranges from northern United States to Brazil but the only Antillean record is incerta Koch from St. Thomas. Menemerus bivittatus (Dufour) = melanognatha (Lucas) is almost cosmopolitan if the svnonymy of Peckham and Petrunkevitch be ac- cepted. Its only Antillean record, however, seems to be Santiago de las Vegas, Cuba. We have it from about plantation buildings at Cabanas, Cuba and also from Culebra. Metacyrba is considered by Simon, but not by Petrunkevitch, to be a synonym of Fuentes. In any case it is solely American, and it belongs to a group, Mzvivee, which is known only from America. The only Antillean species is pictipes Banks from Haiti. Myrmarachne is practically cosmopolitan. 1. melanocephalus Mac- Leay is said by Lucas to be from Cuba and Petrunkevitch so records it without further comment, but the original description says it “is a native of Bengal and I present a figure of it, made by my friend Mr. C. Curtis, in order to show the relation which it bears to the American subgenus, called Myrmecium by Latreille.” M. parallelus (Fabricius) has only the unsatisfactory record of “+Antilles.” Nilacantha cockerelli Peckham is the only species of its genus. It is found in Jamaica and Haiti. The Thiodinz, to which it belongs, is an American group. Peckhamia has but four species. It ranges, on the mainland, from Canada to Panama. A related genus, forming with it the group Peck- hamiez, is only known from Brazil. There are no records from the West Indies except for an immature specimen from Haiti, but Mr. Banks informs me that he has seen a specimen of the United States species, P. picata (Hentz), from Havana and we took an immature specimen of the genus at Cerro de Cabras near Pinar del Rio, Cuba. Pellenes is found in Europe and the Mediterranean region, north- eastern Asia and America. In the latter region there are more than fifty species but it is not known south of Central America. There are — LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 107 but two West Indian records: banks: Peckham and translatus Peckham, both confined to Jamaica. We took specimens of the genus which Mr. Peckham marked as new species at Cristo and Pinar del Rio, Cuba, and we also took coronatus (Hentz) at Pinar del Rio. This species ranges from New York to Mexico. Plexippus is found throughout most of the world but especially in the warmer regions. P. paykulli (Adouin) is a cosmo-tropical species which has been recorded from Santiago de las Vegas and Havana, Cuba; Haiti; Lares, Adjuntas, San Juan and Arecibo, Porto Rico, and St. Vincent. We took it at Guane, Cuba. Prostheclina Keyserling is considered by Simon to be synonymous with Saitis. This is probably correct but as there is some doubt they will be kept separate here except that, in giving the general distribution of Saitis, Prostheclina will be included. In America Prostheclina seems to be limited to the Greater Antilles. P. parvula Banks is known only from Cayamas, Cuba. P. perplexides Strand, venatoria Peckham, and viarva Peckham are known only from Jamaica (+Ipswich, +Port Antonio and +Moneague, respectively). P. morgani Peckham was described from Kingston, Jamaica, and perplera Peckham from Mandeville, Jamaica. Both species are reported also from Haiti. P. illustris (C. Koch) was described from Porto Rico and has since been reported by Banks from an altitude of 2000 feet on E! Yunque in that island. P. signata Banks is listed in the original description from Utuado and Aguadilla, Porto Rico; also from Culebra. He has since reported it from Haiti and we took it at San Turce and Dorodo, both near San Juan, and in the mountains south of Arecibo, Porto Rico, as well as on Desecheo. | Saitis, including Prostheclina, is known from southwestern Europe, western and southern Africa, India, Malasia, Polynesia, Australia and America. There are three species in South America; the others are from the Greater Antilles. With the exception of those mentioned under Prostheclina, all the known Antillean species are confined to Jamaica, namely, anne Cockerell, +Kingston; defloccatus Peckham, +Kingston, and inutilis Peckham, no definite locality given. Synemosyna is an American genus which, with Simonella from the American tropics, forms the group Synemosynex. Two species of Syne- mosyna are found in Brazil, one in St. Vincent, and the United States species, formica Hentz, is listed from Cayamas, Cuba. Thiodina is an American genus which ranges farther north than the other members of the American group, Thiodiner. T. sylvana (Hentz) is known from tSouth Carolina to Panama and, under the name of retia- _ rlus Hentz, from Cayamas, Cuba. 108 ANNALS NEW YORK ACADEMY OF SCIENCES Wala is an American genus. W. peckhami (Cockerell) is known trom +Jamaica and Haiti. We have it from Montego Bay, Jamaica, and Banos San Vincente, Cuba. JW. vernalis (Peckham) has been reported from Santiago de las Vegas, Cuba; Jamaica; Haiti; San Juan and Agua- dilla, Porto Rico; Vieques; +St. Vincent and Bermuda. We took it on Mona and Desecheo; and at Quebradillas, Manati, Dorodo, San Turce and Naguabo, Porto Rico. We also took undetermined specimens of the genus in the vicinity of Pinar del Rio, Vifiales, Cabanas, Cristo and Guantanamo, all in Cuba. Zygoballus and a Brazilian genus, together forming the group Zygo- ballee, are American. It ranges from Panama northward. The only record for the West Indies is swavis Peckham from Mandeville, Moneague and Kingston, t+Jamaica. We have it from Montego Bay, Jamaica. We also have a specimen of the genus, marked by Peckham as a new species, from Cerro de Cabras, near Pinar del Rio, Cuba. DISTRIBUTION OF FAMILIES There are twenty-three families of Araneze now known in the Greater Antilles. Five of these (Dictynide, Cicobiide, Oonopide, Dysderide and Mimetide) are added by this list. One family, Leptonetide, has three species, two genera, on St. Vincent, but it has not yet been found in the Greater Antilles, although it occurs on the mainland from California to Brazil. It is placed by Simon between the Sicariide and Oonopide. Its species are small, long-legged creatures for the most part cavernicolous, and should be looked for in the Greater Antilles. It is fairly widespread but is not yet known from southern Africa, Madagascar, Australia or the southern Pacific Islands. There are eleven small families which are found on the mainland but have not, as yet, been reported from any of the Antilles. The Atypide (p. 77) and Hypochilide (p. 77) have already been mentioned. Zo- dariide is a large and widely distributed family which will probably be found in the West Indies. Senoculide are running spiders found only on the American mainland from Mexico throughout most of South Amer- ica. The Archeide are interesting. There are three species of the type genus, Archea, known from Baltic Amber but only one living species. This species is known only from Madagascar. The only other genus in the family is known by two species from Patagonia. The family Pla- toride has also an interesting distribution,—two genera in South Amer- ica and one in China. The remaining six families (Zoropside, Psech- ride, Eréside, Prodidomide, Zodariide and Hersiliide) have a wider distribution in the Old World. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 109 Of the Antillean families, all are found from north of Mexico to South America except Palpimanide, which in America is known only from South America, St. Vincent and Cuba, and Caponiide, which in America is known only from Brazil to Mexico, St. Vincent, Hispaniola and Cuba (elsewhere only in South Africa). Table I summarizes the Antillean distribution of those families which are known from the Greater Antilles. TasBLeE J.—Antillean distribution of families of spiders : | Dy = | E S 2 ee a a a oa = 3 Sesh oe al hanya. | Se 5 s | VASO 10 © ere x x x .- a ae x Pileiaridee . 2... 2 ses x . < S< = x x ~ Lig Cr x me x x x e Meetipidee- i.e ess x x Sc 4 Myetervd. 52... ss. 4 ie x ar 4 Beem ek >< ae vi 4 Sé 2 0 x al 2s . x Balpamanide..:....-.... Le OS “s ate 2 x WON sk x x x x 4 x WWeridudse....)........ < ySilangd aS x x Si Premppniide. ke. gee adil y yas x x A Pee Sw. < x Sed Ne x x x J) EG a . ae x be on s - MMOMNISIO®...-. .... x ae x x x x Be Brmbennide . se. ye x > eRlls Migiows a x x x 22 5, LG i .- Dee ait eats ip x Ee Lr -- x SE be! bs x 4 x eS es ~< x Si he te x Be Le x x Sr) Me x x 4 21 LS 2S re 4 We SGie abs AS 4 4 x It should be remembered that throughout this paper Lesser Antilles means little else than St. Vincent, since it is the only one of these islands which has been carefully studied. The need of work in Hispaniola is emphasized by the fact that one man collecting for only a few days on the small islands of Mona and Desecheo found representatives of twelve families, while but thirteen families are recorded from the large neigh- boring island of Hispaniola. It is strange that but nine families are known from Jamaica. This island has had the benefit of several workers, but poverty of its fauna is shown in other groups, so that this small num- ber, while certainly not representing all the families to be found there, may be significant. 110 ANNALS NEW YORK ACADEMY OF SCIENCES In view of the careful work done on St. Vincent, it is not likely that its number of families will be greatly increased. It lacks Dictynidex, but this family is in the Greater Antillean list only by reason of a specimen from Guane, Cuba. It lacks Gicobiide, of which we found a species on Mona, but the family is otherwise unknown in the Antilles. Simon be- heves the species we found on Mona to be carried by commerce. It is curious that it should be on Mona, where there is almost no commerce, and not on St. Vincent, where there is a great deal. Mimetide are now known from Cuba and Porto Rico but not St. Vincent. They do not seem to be common even on the first mentioned islands as they had not previously been recorded and we have but one locality for each island. Finally, Pisauride, although not common, are recorded from all the Greater An- tilles and from Vieques but from none of the Lesser. As they are, for the most part, rather large spiders, this is probably not due to defective collecting. It is probably one of those defects in distribution which are likely, when properly understood, to give us most valuable clues to the larger problems involved. DISTRIBUTION OF GENERA As a rule, family is somewhat too large a group to be of much value in a discussion of distribution and species is too small and uncertain, being not only no less human than other taxonomic groups but even more likely to be based on geographic range—the very thing we would like to use it to study. Accordingly genus seems the most favorable group to consider intensively. The two Cohors of Petrunkevitch—Araneus and Dendryphantes—will be left out of consideration and also the following genera: Ischnocolus, Mygale, Blechroscelis, Teutana, Agobardus, Pholcus and Hasarius. The first two are not considered good genera; the next three are recorded from “West Indies” without further data; and the last two are said to be found throughout the tropical countries but no West Indian records are known. LESSER ANTILLES There are 108 genera of which 54 or just 50 per cent. are not known from the Greater Antilles. Seven, or 6.5 per cent., are peculiar to the Lesser Antilles. Ischnothyreus is not known elsewhere in America but is found in western Africa, Ceylon and the Philippines. Likewise, Corin- nomma is known elsewhere only in Africa, eastern tropical Asia, Ceylon, Malasia, Papuasia, Philippines and Australia. In addition to the seven mentioned above there are 40 others which are not found in the Old LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 111 World. In other words, 43.5 per cent. are strictly American. Of these, one, Alcimosphenus, is known from Greater Antilles but not from the continent, 33 are known from South America, 21 are known from Cen- tral America, 14 are known from United States. There are 59 genera which are known in America outside of the Lesser Antilles and are also known from the Old World. One of these, Dry- musa, 1s known only from St. Vincent and Mona in the New World. In the Old World it is known only from Cape of Good ae one species being found in each hemisphere. Of the other 58, 56 (17) are known from South America, 44 (6) are known from Central America, 41 (4) ave known from United States. Leaving out of account cosmopolitan or at least cosmotropical genera and those which are distributed pretty gen- erally throughout the American-mainland, we have the numbers shown in parentheses. Let us now examine the distribution of these more in detail. One of them, Theone, has one species in St. Vincent, one in Missouri, and the genus is represented also in France and Sumatra. Hahnia has several species from Virginia and Wyoming northward to Greenland, two in Patagonia and Terra del Fuego and one in St. Vincent and Porto Rico. It is unknown from the middle of the western mainland. In the eastern hemisphere it has been found in Europe, in the mountains of India, in Japan and Sumatra. Bolostromus is found in Alabama (one species), northwestern South America (four species), St. Vincent (one species) and southwestern Africa. Oxzyopeidon has six species in Central America and Mexico, one of them reaching to Arizona, one species in St. Vincent and Mona, and in the Old World it is found in tropical east Africa, India and Indo-China. One genus found in Central America has just been ait! ‘Phe. remaining five are found in both Central and South America but are not known from United States. One, Beata, is known only from this section of tropical American mainland, St. Thomas, St. Vincent and Africa. The details of the distribution of Miagrammopes, Dysderina, Artema and Stephanopis may be had by consulting the taxonomic part of this paper. One is found in the Mediterranean region, three in Africa, three in Mada- gascar, two in continental Asia, one in the Philippines, two in Malasia, one in Melanesia, two in Polynesia and two in Australia. Seven of the genera under discussion which are found in South Amer- ica have been mentioned. The followimg ten are not known from the American mainland north of Panama and, with the exception of Opopea, which we found on Desecheo, not even from any of the Antilles except St. Vincent: Accola, Theotima, Oonops, Opopea, Janulus, Episinopsis, Dyschiriognatha, Ogulnius, Anapis, Caloctenus. One is found in Europe, 112 ANNALS NEW YORK ACADEMY OF SCIENCES five in Africa, one in eastern Asia, two in Malay peninsula (one not being known elsewhere in the Old World), four in Ceylon (one not being known elsewhere in the Old World), one in New Caledonia, one in Japan, three in the Philippines (two not being known elsewhere in the Old World), two in Malasia and one in Australia. I believe it is impossible to explain these distributions by accidental dispersal in relatively recent times through wind or otherwise or by any recent system of land bridges. Taking into account all the genera found in the Lesser Antilles, we find that 22 (20.4 per cent. of the total) are practically world-wide in their distribution. This leaves 39 genera which have a more or less re- stricted range in the eastern hemisphere. Of these, 7 (18.0 per cent.) are found in central or northern Europe, 10 (25.6 per cent.) m the Med- iterranean region, 23 (59.0 per cent.) im Africa, 5 (12.8 per cent.) m Madagascar, 20 (51.3 per cent.) in continental Asia, 10 (25.6 per cent.) in Ceylon, 8 (20.5 per cent.) in Japan, 8 (20.5 per cent.) in Philippines, 12 (30.8 per cent.) in Malasia, 5 (12.8 per cent.) in the smaller Pacific Islands, and 8 (20.5 per cent.) in Australia. The American distribution outside of the Lesser Antilles of Lesser Antillean genera is shown in Table II. The study of this table had better be deferred until after the Greater Antilles have been considered. Taslte I].—American distribution of Lesser Antillean genera of spiders* Not | on | SAL) (A OS. 8 As) eae a - main- only only only CA. | U.S. | U.S. |) 283 eee land | Not in other An- pti |= eee Spey eae ee 9 19 | # 1 6 2 1 14 5+ Bo Henly 5:5 1 1 2 1 1 7 Hisp. only. ..... 1 1 2 Cuba only...... 3 i 2 1 3 10- Jamm..; qubyn.3..'.' 1 1 Po wap. 2 32 1 L P. R., Cuba a age ae 1 as 3 + Hisp., Cuba.... es = = 5c. Bt ee. rs = =e Z ai 2 Leas. Pele 1 zi 3 - Tess... Ehispe 2 24). 1 3 - PL hi. Jam | | me 1 1 Less., P_E., Hisp- | | 3 9 12 Less., P. B., Jam. VAs nh 2 2 Less., Hisp., Jam. © 19) if | ; 2 P.R.,Hisp, Jam. | .. | | | 5 5 Less., P.R., Hisp., | | Page Bok oe Ah 1 1 6 8 ital e 5. es RE BS 6 9 5 4 | 49°) ‘Bf | JAMAICA Only 31 genera of spiders are known from Jamaica, including Pros- theclina. Of these 7, or 22.6 per cent., are not found in the other An- tilles. The question as to whether Prostheclina of Keyserling should be treated as a synonym of Saitis of Simon has been mentioned several times and Prostheclina has been omitted from the discussions when world-wide 70 Columns refer to the mainland and rows to the Antilles. ‘“‘C. A.” includes Mexico, and “U.S.” everything north of Mexico. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 119 distribution was concerned. In Jamaica we are troubled by the fact that three species of Saitis are recorded. There are three other species in South America and the genus in the broad sense is known from south- western Europe, western and southern Africa, India, Malasia, Polynesia and Australia. If we include the records for Prostheclina with Sastis the result will be to decrease the apparent distinctness of Jamaica from the other Antilles, but as it is still quite distinct this would probably be the fairest course. We then have 30 genera credited to Jamaica of which 6, or 20 per cent., are not known from the other Antilles. Of these 3, or 10 per cent., are ‘not known elsewhere in America. Theridionexus and Epecthinula are confined to Jamaica and Cyatholipus has been found elsewhere only in South Africa. Fourteen genera (43.3 per cent.) are strictly American ; 10, or 33.3 per cent., are practically cosmopolitan or cosmotropical. In addition to these and Sastis and Cyatholipus, which have just been met- tioned, Pachylomerus is known from Jamaica, St. Vincent, Brazil (?), Venezuela to Maryland, the Mediterranean region and Japan. Ischno- thele is found in Jamaica, South America to Mexico, Bahamas and either it or a closely related genus, which we will consider as it, in eastern Africa, Madagascar and India. Filistata is known from South America, including the Galapagos, to California and southeastern United States, Bermuda and the Atlantic islands, Mediterranean region, Africa, central Asia, Philippines and Australia. Pellenes is reported from Jamaica, Cuba, Central America to Canada, Europe, Mediterranean region and northeastern Asia. The numbers are so few that the reader can easily analyze the distributions from the data just given. Of the 10 American genera which have been found on the mainland, 6 are known from South America, 10 are known from Central America and Mexico, 7 are known from United States. The American distribution of Jamaican genera is shown in Table VI. 120 ANNALS NEW YORK ACADEMY OF SCIENCES TaBLeE VI.—American distribution of Jamaican genera of spiders™ Not S. A on Se Ae Ce U.S S. A WA. 12s. As. eee main-| only only only. | CA. | U. S._|-U. 8. | USSetene land Not in other An- tillese = 3 ae 3 at 1 oe cL 1 6 lesser only.2--- ee 1 it Hisp: -onlysa2e 1 I Cuba only. 4-2. : 2 2 ess. akuispee wake Bek Se oy ate ay 1 ie eee is 1 PR; Cubase. = Bt ae bie ae a ee if 1 Less., P.R., Cuba a ee ig se a — 2 2 Less., Hisp.,Cuba 1 - se ie if 3 ee x 2 P. R., Hisp., Cuba Bis 1 oh a af = sR ) 6 Less., P. R., Hisp., Guibas. cere se ne iJ Wee i ie a4 6 8 Totals pcre 5 at 2 0 3 3 0 16 30 GREATER ANTILLES Omitting the two cohors and the seven doubtful genera mentioned on page 000; also considering that Misumessus echinatus is a Misumenops. that the Antillean records for Prostheclina in Petrunkevitch’s Catalogue should be credited to Saitis and that Dolomedes has not been recorded from Cuba, there are 117 genera left, of which 63, or 53.8 per cent., are not known from the Lesser Antilles. Six, or 5.1 per cent., are peculiar to the Greater Antilles. They are Hapalopinus in Cuba, Scopelobates in Hispaniola, Theridionexus and Hpecthinula in Jamaica, Bythocrotus in Hispaniola and Nilacantha in Hispaniola and Jamaica. Cyatholipus is not known in America outside of Jamaica but occurs in South Africa. in addition to the six mentioned above there are forty-four others which are not found in the Old World. In other words, 42.7 per cent. are strictly American. One of these, Alcimosphenus, is also known from the Lesser Antilles (St. Vincent) but not from the mainland; 33 are known from South America, 30 are known from Central America and Mexico, 24 are known from United States. There are 66 genera which are known in America outside of the Greater Antilles and are also known from the Old World. One of these, Drymusa, is known only from Mona, St. Vincent and Cape of Good Hope. Of the other 65, 58 (9) are known from South America, 56 (7) are 1 Columns refer to the mainland and rows to the Antilles. ‘“‘C. A.” includes Mexico, and “U.S.” everything north of Mexico. ~~, A LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 121 known from Central America and Mexico, 55 (7) are known from United States. As before, the numbers in parentheses refer to genera of more or less restricted distribution. J/ysmena, Ceratinella and Hyctia are Holarctic genera which are for the first time in West Indian list and now only by reason of being found in Cuba. The others (/schnothele. Miagram- mopes, Dysderina, Opopea, Artema, Stephanopsis, Syrisca, Hahnia, Oxyo- peidon, Pellenes and Saitis) are considered rather fully in the discussions concerning the individual islands and also in the taxonomic part of this paper. | Of the 117 genera in the Greater Antilles, 34 (29.1 per cent. of the iotal) are practically world wide in their distribution. This leaves 33 genera which have a more or less restricted range in the eastern hemi- sphere. Of these 10 (30.3 per cent.) are found in central or northern Europe; 13 (39.4 per cent.) in the Mediterranean region: 22 (66.7 per cent.) in Africa; 7 (21.2 per cent.) in Madagascar; 23 (69.7 per cent.) — in continental Asia; 3 (9.1 per cent.) in Ceylon; 7 (21.2 per cent.) in Japan; 6 (18.2 per cent.) in the Philippines; 8 (24.2 per cent.) in Malasia; 5 (15.2 per cent.) in the smaller Pacific islands and 9 (27.3 per cent.) in Australia and New Zealand. The American distribution of genera found in the Greater Antilles is shown in Table VII and those found in the Lesser Antilles are added for comparison. TABLE VII—American distribution of Antillean genera of spiders ® Greater Totals Greater and Lesser ps only Lesser only Greater Lesser W. I. Not on mainland... rs 2 9 9 af 18 Pee Only.....2... 5 5 19 10 24 29 2 2 af if 3 2 = S800 ~ 2 1 6 3 7 oo 6 10 6 16 16 22 o 2 6 2 at 8 3 9 - 2 ee 4 2 3 6 5 9 ooo. A. U.S... 29 30 14 59 44 yes SS es Sa ee ——— ee SS _ ere 63 54 54 117 108 EYAL: 2 «“C. A.” includes Mexico, and “U.S.” everything north of Mexico. 122 ANNALS NEW YORK ACADEMY OF SCIENCES DISTRIBUTION OF SPECIES Table VIII summarizes the data at hand. A few doubtful species have been omitted and several undescribed ones have been included. TaBLeE VIII.—American distribution of Antillean species of spiders *® Not S.A on S. A Ch: U.S S. A C. Ac Sa Cie main- only only only CUA We. es U.S Total land Lesser only..... 88 18 2 2 4 2 di ~ 121 P. Be enkys 2 22 6 2 : aye 2 1 1 12 Hasp: only. = 3:7 18 if 1 2 1 4 1 28 Cuba Uonlvese.-- 33 2 3 22 se § i 5 74 Jaime, Orby. oo. 23 ] 1 1 26 Less PR oe 9 1 1 bs 11 Less.,-Hispooc: 1 ae 1 2 Bess: “Cuban a 3 ee 2 i 2 8 PRS isp. es 2 2 a 2 oe 7 PL i Cabae 2 2 1 2 2 9 Hisp. Cuba: ot: 5 2 3 10 Easp., Jam eee, 5 5 Guba, Jam zc... 1 ms 1 Less., P.R., Hisp. 2 1 es 3 Less., P.R., Cuba 2 1 i 2 6 Less.. Hisp., Jam. 1 1 P:R” (rash. Saat es ae “e 1 1 Hisp., Cuba, Jam. il i Less., P.R., Hisp., Gubay see 2 af 2 2 2 9 Less., P. R., Cuba, a Ih nen os ee ss 2 oe ae = 2 2 ey eY) 5 fH = ate 2. Eee ee i ge ee yes 2 md 2 5 All (2) Antilles. 1 3 4 Meped. oe 206 |" 2a) a0 A aoD red a emo 6 | 30 | 347 GENERAL DISCUSSION Is THE FAUNA OF THE LESSER ANTILLES DISTINCT FROM THAT OF THE GREATER ANTILLES ? Comparing the two usual divisions of the Antilles—Greater and Lesser—we see that 50 per cent. of the Lesser Antillean genera of spiders, unfortunately meaning little more than the fauna of St. Vincent, have 78 Columns refer to the mainland and rows to the Antilles. “C. A.” includes Mexico, and “‘U. S.”’ everything north of Mexico. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 123 At rot been found in the Greater Antilles and 53.8 per cent. of those known from the Greater Antilles have not been reported from the Lesser An- tilles. In the Lesser Antilles 6.5 per cent. of the genera are peculiar and in the Greater, 5.1 per cent. In the Lesser Antilles 43.5 per cent. are strictly American and in the Greater, 42.7 per cent. The last two comparisons indicate considerable similarity in the general character of the two faunez and raise doubt as to the significance of the fact that half of the fauna of each division is distinct from that of the other division. Do the two divisions have significantly distinct faune? Taking His- paniola because it is in the midst of the Greater Antilles and bearing in mind that further study of the Hispaniolan spiders is more likely to bring out differences than similarities because we are now ignorant of the fauna of the interior, we note that 42.3 per cent. of its genera are not known from the Lesser Antilles, 44.3 per cent. are not known from Porto Rico, 30.8 per cent. are not known from Cuba and 65.4 per cent. are not known from Jamaica. The case of Jamaica is probably due, in large part, to ignorance and a further study of Cuba and Porto Rico without a further study of Hispaniola would doubtless reduce their fig- ures, but it must also be kept in mind that data for the Lesser Antilles is largely drawn from an island well down the line, so that we must con- clude that with respect to the genera of spiders the Lesser Antilles are no more different from the Greater Antilles than the different islands of the latter are from each other. Of course, species show a much more limited geographic range than genera and the actual distance of St. Vincent from the Greater Antilles apart from purely faunistic factors becomes an important factor when considering species. ‘There are 168 species known from the Lesser An- tilles of which 72.0 per cent. are not reported from the Greater Antilles, and of the 226 species known from the Greater Antilles 79.2 per cent. are not known from the Lesser. In the Lesser Antilles 52.4 per cent. of the species are not known elsewhere and in the Greater Antilles, 42.5 per cent. It will thus be seen that the spider fauna of the rather isolated Lesser Antilles is not much, if any, more specialized than that of the Greater. Again taking Hispaniola for a basis of comparison we find that of its 76 species, 73.7 per cent. are not known from the Lesser Antilles, 68.4 per cent. are not known from the Porto Rican islands, 60.5 per cent. are not known from Cuba and 77.6 per cent. are not known from Jamaica. To say the least, the material at hand would not justify us in considering the Lesser Antilles any more distinct from the Greater than any of the islands of the Greater are from each other. 124 ANNALS NEW YORK ACADEMY OF SCIENCES Is THE ANTILLEAN FAUNA DISTINCT FROM THAT OF THE MAINLAND? The introductory paragraph to Scharff’s Chapter on the origin of the West Indian fauna is as follows: “North and South America are to be regarded. according to Professor Suess, as two essentially distinct land-masses, between which is interposed, as a third element, the area of Central America and the Antilles. This geological dis- tinctness of Central America and the Antilles from the two neighboring conii- nents is scarcely recognizable in the fauna of the great isthmus. But the West Indies are comparable to a wedge driven in between two faunistically, more or less, independent and distinct masses. Almost everyone who has dealt with the fauna or flora of the West Indian islands has expressed his surprise at this fact. In position. says Dr. Wallace, the Antilles form an unbroken chain uniting North and South America, in a line parallel to the great Central American isthmus. Yet instead of exhibiting an intermixture of the produc- tions of Florida and Venezuela, they differ widely from both these countries, possessing in some groups a degree of speciality only to be found elsewhere in islands far removed from any continent.” Several years ago (1913) I published a brief note on the distribution of occidental spiders, getting the data entirely from Petrunkeyvitch’s catalogue. As has been done here, I roughly divided the western hemi- sphere into four parts and found that 59.8 per cent. of the South Ameri- can genera of spiders were not known elsewhere in America, 21.5 per cent. of those in Central America and Mexico were peculiar to that region and 37.5 per cent. of those from United States and Canada were not found farther south. Of the 133 Antillean genera then considered 13.6 per cent. were known only from the West Indies. The present paper includes 171 genera without adding any peculiar ones, so that the per- centage is reduced to 10.5. This would also reduce the percentage for some of the other divisions but the figures as they stand are about as reliable as the data on which the revised ones would be based, genus being an indefinite sort of thing, so that we may conclude that the genera of spiders do not confirm the idea of the distinctness of the West Indian fauna. Instead of being very distinct from the mainland they are less so than any of the mainland divisions are from each other. Is this be- cause spiders are more generally distributed than other animals or is the impression of distinctness an erroneous one based on a consideration of special cases ? The best way to settle the question of the distinctness of the Antillean fauna is to study carefully a number of groups which are fairly well represented in the Antilles and we are doing this in connection with the survey of Porto Rico which is being carried on under the auspices of the New York Academy of Sciences. Mammals were used by Wallace with LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 125 wonderful success but they do not seem to be good material for a West Indian discussion because there are so few of them. It is decidedly in- teresting that, except for bats, the Antillean mammals are rodents be- longing to South American groups and Solenodon, a genus of Insectivora, whose nearest relatives are found only in Madagascar. ‘These relation- ships will be referred to later but it may be noted in passing that such distributions are fairly common among spiders. ‘The figures given by Wallace for birds show that about one-third of the Antillean genera are peculiar. This proportion is far greater than that for spiders and, when Fie. 3.—Sandy plain south of Pinar del Rio, Cuba compared with the bird fauna of Central America, for example, shows a high degree of speciality, but Wallace’s list gives only 95 genera of birds for the West Indies and there are more than 500 genera in Central America. The small number of vertebrates in the West Indies and the fact that a large proportion of those which are there are peculiar seem to be evidences of the unfitness of the Antillean environment for yverte- brate life rather than of any special distinctness of the Antilles from the standpoint of geographic distribution. It seems almost pedantic to point out the necessity of considering en- 126 ANNALS NEW YORK ACADEMY OF SCIENCES 92Taphic distribution and vet it is almost @ never taken up in detail. There are several very practical reasons for eae : ss a 7 bere Pg el ee ene ee are Well liiustrated DV this paper. First. there 1s Lenadis: cia a, See ie lee ae a 7. ee aiready Such a Mass Of detall that 1t Can scarcely be nandied and. second, py Lad pe) eee Sy Ca ee oe 2 en Ps > the author is oiten, as 1m this case. ignorant of the ecological conditions under which the species occur throughout their ranges. The importance of ecological conditions may be illustrated in connection with what has = = = . 7 s tT just gone before as well as with what follows 1G. 4—View near Banos San Vincente. Cuba The level area is rich agricultural land. and even the cliffs support a rather luxuriant vegetation. Mr. Leng and myself went to Pifar del Rio, Cuba, chiefly to study the ecological distribution of insects. We collected spiders in a number of environments, two of which are of interest here. South of the city is a sandy plain which we studied fairly intensively between the nime and thirteen kilometer posts on the road to Coloma. Leaving out of account the really important local differences caused by ground water level and the resulting differences in vegetation, figure 3 may be taken as fairly typical of the region. North of the city are the mountains and we col- LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 127 lected especially in the vicinity of Bahos San Vincente. ‘The difference in elevation between this place and the plains is not enough to cause a noticeable difference in temperature apart from that connected with soil conditions and moisture. Figure 4 gives an idea of the character of the region—moist atmosphere, lime-humus soil, rather “tropical” vege" tation. Leaving out of account the Cohors Araneus and Dendryphantes for taxonomic reasons and not considering the minor ecological differ- ences in each region, we found 10 genera on the plains which we did not find at Banos, 24 at Banos which we did not find on the plains and only 9 genera at both places. That is, 52.6 per cent. of the plains genera were not found in the hills and 72.7 per cent. of the hills genera were not found on the plains. In other words there is at least as much differ- ence between these two nearby but ecologically different localities as there is between Cuba and the other islands or between the Antilles as a whole and the mainland. Furthermore, this must give us a pause in considering the relationships of the various islands to each other and to the mainland, for 90 per cent. of the genera we took on the plains are found in United States and only 77.3 per cent. of those from the hills; 40 per cent. of the former are known elsewhere only in United States, as far as America is concerned, and only 4.5 per cent. of the latter. The general impression is that the West Indies are more closely related to Central and South America than to United States and, going still further, that the Lesser Antilles are especially related to South America and the Greater Antilles to Central America. If this be true, in how far is it due to land bridges, ocean currents or wind and in how far is it due merely to similarity, and hence congeniality, of environment? ‘The first part of the question needs prior consideration, and the second part cannot be satisfactorily answered until we know more about animal ecology. MAINLAND AFFINITIES OF THE ANTILLEAN FAUNA As far as the data concerning the genera of spiders go, the Lesser Antilles (St. Vincent) seems to have a closer affinity with South America than does the other division of the Antilles, since 22.2 per cent. of the genera of the Lesser Antilles are known elsewhere in this hemisphere only in South America and the corresponding percentage for the Greater Antilles is only 8.6. However, if we include all the genera whose dis- tribution we can study we note that only 82.4 per cent. of the genera known from the Lesser Antilles are known also in South America, while the corresponding percentage for the Greater Antilles is 84.3 per cent. The belief that work in the northern Lesser Antilles will still further ANNALS NEW YORK ACADEMY OF SCIENCES oa) 12 reduce the proportion of strictly South American genera in that division makes us somewhat dubious of there being any marked difference between the two divisions of the Antilles with respect to this relationship. As far as the affinity between the Greater Antilles and Central America (including Mexico) goes, we note that only 2.6 per cent. of the Greater Antillean genera are confined, in their American mainland distribution, to that region. Including all the genera, 73.5 per cent. of those known from the Greater Antilles are known also from Central America and Mexico, but the same percentage for the Lesser Antilles-is 60.2, which is practically the same, in view of the fact that it is so largely based on the St. Vincent fauna. As was pointed out several paragraphs before, Cuba has a number of genera known on the American mainland only from United States and Canada, but such genera represent only 5.1 per cent. of the genera known from the Greater Antilles (as a matter of fact, only 7.4 per cent. of the Cuban genera), while the same percentage for the Lesser Antilles is 2.8. Of the Greater Antillean genera 67.5 per cent. are known also in United States and Canada and 50.9 per cent. of those reported from the Lesser Antilles. This, it seems to me, confirms the notion that, while individual genera differ, the general make-up of the spider fauna of the Lesser Antilles does not significantly differ from that of the Greater Antilles and that neither division has drawn from, or given to, any particular portion of the mainland much more than the other division. A comparison of the Porto Rican spiders with those of the other islands brings out some interesting points. It has, as far as we know, no peculiar genera and only 9.8 per cent. of its genera are not known from the other. Antilles. The data for the Lesser Antilles show 6.5 per cent. of peculiar genera and 50 per cent. which are not known from the other Antilles: for Hispaniola the figures are 5.8 per cent. and 17.3 per cent.: for Cuba, 1.2 per cent. and 29.5 per cent.: and for Jamaica, 6.6 per cent. and 20 per cent. The fauna of Porto Rico seems, then, to be of more recent origin than that of the other islands. This idea is strengthened by the fact that, while the differences are not great, only 30 per cent. of Porto Rican genera are strictly American as compared with 43.5 per cent. for the Lesser Antilles, 47.1 per cent. for Hispaniola, 38.5 per cent. for Cuba and 43.3 per cent. for Jamaica: also by the fol- lowing, which is not strictly a restatement of the preceding, since a genus may be found in the Old World without being so widely distributed as to be classed as cosmopolitan. Cosmopolitan or cosmotropical genera make up 44.0 per cent. of the Porto Rican genera as compared with 20 LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 129 per cent. for the Lesser Antilles, 41.2 per cent. for Hispaniola (practi- cally, the port towns), 35.8 per cent. for Cuba and 33.3 per cent. for Jamaica. Species also show the composite character of the Porto Rican spider fauna and are perhaps more convincing than genera. Only 8.7 per cent. of the Porto Rican species are peculiar as contrasted with 52.4 per cent. for the Lesser Antilles, 23.7 per cent. for Hispaniola, 25.2 per cent. for Cuba and 48.9 per cent. for Jamaica. In view of this unusual character of Porto Rican spiders, it is worth while discussing their distributicnal affinities. Table VIII gives the deta for the American distribution of species in a condensed form. Table IX shows the percentages of the total number of species in the several islands for various groupings. (ABH bx Lif. PR: | Hisp. | Cuba. Jam. phen s | ee ee eee SEA. 52.4 Bia | 23.7 | Baa ARS Mig eOiLHer ANGtIIEGS. 2.2. 62 ess tere oc | 1220 gO: a Wey oa | 56.5 Hots In other Antilles but not on mainland. | 13.1 29.0 A276 ESE SBS: On mainland but not in other Antilles. | 19.6 Sau A | ald 6:4 mantic ( tObal) 20.3... ke eae ess | 34.5 62.3 AB OF hd oh. 29.8 Porto Rico has as high a mainland affinity as Cuba and higher than the other islands, but it has a lower direct affinity (a lower percentage of species found on the mainland and not on the other Antilles) than any except Jamaica. That is, its fauna is largely mainland species which it has received by way of the other islands; Cuba’s species are also largely mainland species, few of which have been passed on, while Jamaica has few mainland species and most of these have been received by way of the other islands or have been passed on to them. Of course, it would be possible for a species to originate in a West Indian island and then be transferred to the mainland and this may be true in Cuba. It would be impossible to determine from the data whether this has hap- pened or not but it probably did not occur often enough to seriously complicate matters. The rest of Table 1X agrees with the idea just stated and repeats what was said before, as might be expected, since it is largely a complement of the two lines just considered. Porto Rico has the smallest percentage of endemic species of any of the islands or groups of islands; and, partly for this reason, but partly also because it has received a large proportion 14 Hor explanation, see text. 130 ANNALS NEW YORK ACADEMY OF SCIENCES of its species from the mainland by way of the other islands, it has the smallest percentage of species which are not in the other Antilles. Porto Rico, Hispaniola and Jamaica contrast with the Lesser Antilles (St. Vincent) and Cuba in having a greater percentage of species which have criginated in the Antilles and been passed to each other, but not to the mainland. Table X gives the percentages of the species occurring in a given island and also on the mainland which are found in the several mainland divis- ions. It is an attempt to discover the way species have moved in rela- tively recent times. It probably concerns relatively recent movements only, for the species have not changed enough to lead taxonomists to give the mainland and insular lots different names. ‘T’o be sure, we know that specific identity may be preserved for long geologic time and TABLE X ¥ | IL@Seeie |) 1B, Les Hisp | Cuba Jam i Ul arene ne ee SS SP i | Su AL omy. gg es Pee dee 3160 11.26 10.8 3.8. |) f4ss Ch Aa SOM Sao ahs eee aan 5.2 (ea0) 2.7 6.3 fo Ue Si Ory hee erie 6.9 e.) 10.8 30.0 0.0 Sort AL EG 3 As TIE eee aes 2 14.0 8.1 6.3 how Cis, SANT BS ards tk OIE a ie 8.6 20.9 29.7 PA es 14.3 Saas WesS ec cee wee tte Bree 720 ie: ee 5 0.0 SP A CT IS INES ee ocn oie 2 Or gu acre 32.4 30.0 aT 2 Tore: eee 79 30 |. Goa 56.8 42.5 78.6 taal "Ge AN ite. ge ete DSO ie aye ee (220 a) Bears Sal7 Matyi; Auli keer 46.5 67 .4 18230) 1 Sas TAA. | experimental work has indicated that a species may arise not only sud- denly but in several quite independent centers. However, most of the species considered here probably arose in relatively recent geologic time and each in a single center. Jamaica is troublesome because 33 (all but 14) of the species known from there do not occur on the mainland. Cuba has 51 such species; Hispaniola, 39; Porto Rico, 26, and the Lesser Antilles 110. Except for the strictly South American species in St. Vincent and the strictly northern species in Cuba a large part of the species which are found in the various Antilles and also on the mainland are wide ranging on the mainland, being found in all three of its divisions. This may be due to the greater chance they have of getting to the Antilles (supposing they originated on the mainland) because their mainland distribution 15 Hor explanation, see text. Based upon 58 species in the Lesser Antilles, 43 in Porto Rico, 37 in Hispaniola, 80 in Cuba, and 14 in Jamaica. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 131 puts them on three sides of the Antilles or, as 1s more probable, they are species which can live in a wide range of environments, as is indicated by their wide geographic range, and so have found it easy to establish themselves in the Antilles. The Central American affinity is really slight. Although in the totals it seems to be important, an examination of the more detailed part of the table shows that this is caused by the groups in which it is joined with either the northern or the southern division or with both. For example, the 17.2 per cent. of the mainland species in the Lesser Antilles which are credited to South America and Central America should probably be eredited to South America as is indicated by the fact that 31.0 per cent. are known on the mainland only from South America. The 5.2 per cent. which are known from the mainland only in Central America are three species which may have come directly from Central America but they more probably occur in South America or did recently occur there. Jamaica is the only island with a greater percentage of species which, as far as we know, are strictly Central American, than it has of species which seem to be confined to one of the other mainland divisions, but this percentage is misleading also as it represents but one species. Presumably these species have reached their island homes from the mainland or have originated in the islands and spread to the mainland since the various islands have been separated, if, indeed, the islands were ever joined. The fauna of the islands at the ends of the Antillean chain have their strongest affinities with that part of the mainland which is nearest to them partly because of their geographical contiguity but partly also because of resemblance of habitat, witness the Floridian affinity of the fauna (and it is true also of the flora) of the sandy plains near Pinar del Rio, Cuba, as compared with the more tropical character of the moun- tains to the north of the city. If these species have spread themselves in this way without the aid of actual land connection, is such connection indicated in the older units, the genera ? Jt is doubtless clear that if we take all genera and treat them as we have just treated the species the result would be influenced by the in- clusion of recently introduced species. Probably—but certainly not nec- essarily—those genera which have species peculiar to the Antilles have been on the Antilles longer than those whose only Antillean species occur also on the mainland. Table XI gives the percentages of such presum- ably old genera occurring in a given island and also on the mainland which are found in the several mainland divisions. It is an attempt to discover the way the fauna moved in relatively remote times. 132 ANNALS NEW YORK. ACADEMY OF SCIENCES SABLE: DX Lesser 12 deve Hisp. Cuba Jam. S.A. only ee oe eee oil 6.9 9.4 10.2 5.9 Cy Ag tomy econ eee 2..8 3.4 6.3 4.1 Judi U.S: sonlyes Se ee eee 4.2 3.4 0.0 Oak 0.0 S.. A. GC. DAC ee ee eons 21.1 ies DAR 18.4 11.8 CAL CUS ene errs 4.2 3.4 3.1 4.1 eT Sa Ass SUNS hose ee ee ee ae 7.0 3.4 6.3 6.1 0.0 S. Ac Vel NeW Son een 39.4 62.1 53.1 51.0 53.0 otal) SE VAC S a ates 88.8 89.7 90.6 85.7 70.6 etal Gi Asks nc aeons 67 .6 86.2 84.4 tage 94.1 TotaleoUilesrs ss0 eee ae 54.9 72.4 62.5 3 (0.6 As with species, a large part of the genera which are found on the various Antilles and also on the mainland are wide ranging on the main- land, being found in all three of its divisions. A part of this is, of course, due to those genera containing wide ranging species which also have species peculiar to the Antilles. We also note that, as with species, the strongest generic affinity of St. Vincent is with South America and probably most of the 15 genera (21.1 per cent.) found in both South and Central America got to or from Central America by way of South America. It has already been pointed out that the Porto Rican spider fauna is largely recent. The data for genera as well as that for species indicates that its affinity is probably strongest with South America. It would not be profitable to discuss Hispaniola at length until we know more of the fauna in the interior, but it is interesting to note that no strictly United States genera have been found, and it is safe to predict that further study of Hispaniola will show it to be even more South American than is Porto Rico. Cuba has a surprisingly small percentage of United States genera if we leave out of account the wide ranging ones. The facts that 30.0 per cent. of the species which are found on the mainland are not found on the mainland south of the United States as compared with 6.1 per cent. of presumably old genera, and that the figures for those not found north of South America are 3.8 per cent. and 10.2 per cent. respectively indi- cate that the older fauna has a stronger South American affinity than the more recent. The meaning of this will be discussed presently. There are only 17 genera in Jamaica upon which to base the percentages given in Table XI. A strong affinity with Central America in ancient 16 For explanation, see text. Based upon’71 genera in the Lesser Antilles, 29 in Porto Rico, 32 in Hispaniola, 49 in Cuba, and 17 in Jamaica. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 133 times is indicated, but as it is based on but 2 genera found on the main- land only in Central America and Mexico, 2 found in these regions and South America and 3 found in these regions and United States, it would not be safe to place muth reliance on these data. Furthermore such affinity does not demand a land bridge; it may merely be a result of geo- eraphic contiguity and environmental similarity. If Jamaica has always been isolated as it is now, or at least if its last complete submergence was a long time ago and it has since been as isolated as it is now, we can understand the apparent poverty of its fauna, the peculiarity of it and its only shghtly stronger affinity with Central America than with the other mainland divisions. The impression all these things make on me is that the various An- tilles may always have been as distinct as they are now; that they re- ceived their spider fauna by slow “accidental” means, and that Porto Rico has only recently been populated. How, then, are we to explain the ancient character of a large part of its fauna and the curious relation- ships with distant Old World localities? Also, why is the ancient portion of the Cuban fauna more South American than the recent? ORIGIN OF THE ANTILLEAN FAUNA Before considering further the origin of West Indian spiders it would be well to note their means of dispersal. Young spiders of nearly, if not quite, all families are more or less given to “ballooning.” They will face against the wind, and, elevating their spinnerets, spin a quantity of fine silk usually in the form of a number of threads. These threads float in the breeze and finally they are numerous enough or long enough to carry their makers on an aerial journey. The length of such journeys would seem to depend very largely on the strength and character of the wind. McCook has attempted to show by its distribution that Heteropoda vena- toria has circumnavigated the globe. It is true that the range of this species corresponds “with remarkable exactitude” to the belt over which the Trades blow, but we may accept his proof that the species has not been distributed by commerce without adopting his suggestion that its cosmotropical distribution is due to its ballooning habit. He cites Dar- win’s note about the “Beagle” being boarded by “flying” spiders when sixty miles from land and adds a report which is similar except that the latter ship was more than two hundred miles from land. It must be ad- mitted that wind may be a very efficient factor in the distribution of many organisms, and it should be noted that when spiders go on such journeys they often go in. swarms, so that it would not be unlikely that opposite sexes would land near enough each other to continue the species. 134 ANNALS NEW YORK ACADEMY OF SCIENCES However, there is more to the distribution of spiders than wind. We need an explanation of the remarkable discontinuities which have been pointed out. Ocean drift is frequently brought in to explaim distribution and it has probably been the effective agent in many instances. The opponents of such an idea forget the long ages in which accidental drift has had a chance to work. However, the ease with which large numbers of spiders take to the air every year makes recourse to the small numbers that may make successful voyages on driftwood unnecessary. Then there is man. Hardly a ship sails from port without araneid stowaways, and inland shipments of freight also carry their quota, but these are nearly always a certain few species of which only those that live about man’s dwellings are likely to become established. None of the “tarantulas” which come in nearly every shipment of bananas have be- come a part of the New York fauna. Furthermore it is very unlikely that the distribution of most spiders has changed much since man began to sail the seas. The study of their movements must go far back of that. One takes up the question of land bridges with something akin to a groan since opinion on the matter is so diverse and the evidence, whether pro or con, about a given bridge is often so slight. Paleontology offers little direct evidence as to the ancient movements of spiders since so few fossil spiders are known. Several have been described from the Carbon- iferous of both hemispheres and probably at least one genus (Arthroly- cosa) was even then found in both Europe and North America. The only living genus of the type of spiders which was apparently common in Carboniferous is Liphistus. It is found now only in the islands of Pinang and Sumatra. Spiders with unsegmented abdomens, that is all living spiders except Lipistius, may not have arisen until the Mesozoic, but, if so, evolution was fairly rapid, for most of the Oligocene and Miocene spiders belong to present-day families and even genera. R. T. Pocock** gives an interesting analysis of the ancient movements of spiders. A few of the present-day distributions given by him differ from those given by other authorities, but none of the differences which I have noticed would materially change his argument. In his section on the “Distribution of some of the Families of Arachnomorphe that were represented in the Oligocene Period” he mentions seventeen genera found in amber which are still hying. While it is perfectly true that “although since the Oligocene these Spiders have had the same time for dispersal, they nevertheless differ greatly in their distribution,” the implication that Proc. Zool. Soc. London, 1903, I. pp. 349-568. LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 135 their present distribution is a measure of their success in extending their range does not necessarily follow. It is altogether probable that spiders fairly covered the earth in Oligocene times, and the few that were im- prisoned in Baltic amber tell us nothing more than that they lived in that region, among others, at that time. It is pleasing to note that Pocock did not think it necessary to throw a bridge from continent to continent on the shortest line (according to Mercator’s projection) between any two portions of a present-day discontinuous range. He naturally found con- siderable evidence of transfer in Arctic regions** and he also used, rather freely, Antarctic connections between South America, Africa and Aus- tralia. F. Dahl*® has considered the distribution of spiders and concluded that the Antarctic connection is improbable. Not all of his arguments seem to be well founded. For example, one can not consistently uphold the “Telict” idea and then combat the Antarctic connection on the ground that so few relicts are found on the Antarctic islands. Most of the spi- ders, for the explanation of whose distribution an Antarctic connection might be desired, are tropical. It is therefore not surprising that re- licts are not found on Antarctic islands. Ité is no more than ordinary common sense to favor the simplest ade- quate explanation of a problem. It is unnecessary to review the volu- minous literature concerning Antarctic connections between the three southern continents. It is admitted that many facts favor such connec- tions, especially the one between South America and Australia, and that nothing has been, or is likely to be, brought out which will absolutely disprove it. However, if a simpler explanation than the general eleva- tion of more than 3,000 meters required to connect South America and Australia, but which leaves Africa still to be accounted for, is adequate it would seem to be preferable. The final court of appeals is, of course, fossils, especially those in Antarctica, but even if numerous fossils are found in Antarctica and they are seen to be similar to those found in the southern portions of the other continents it will not prove actual con- nections. It will merely show that there has been an interchange of fauna—a thing not at all unknown between absolutely unconnected land areas and a thing for which there was a vastly longer time than we ordinarily have in mind since it requires a strong effort for us to bring ourselves to thinking in terms of millions of years. 18 The theory that practically all Tertiary migration was by way of the Arctic regions has been set forth most clearly by W. D. Matthew. It is amplified and convincingly dis- cussed in his recent paper on “Climate and Evolution,’ Annals N. Y, Acad. Sci., Vol. AXIV, pp. 171-318. 1915. *% Zool. Anzeiger, vol. XXXVII, pp. 270-282. 1911. 136 ANNALS NEW YORK ACADEMY OF SCIENCES In addition to, and partly because of, our relative ignorance of ancient spiders, we known but little of araneid phylogeny. We do know that spiders were already fairly well distributed, at least in the northern emisphere, in the Carboniferous. Modern families and even genera were well differentiated in Oligocene. Therefore, as far as distribution problems are concerned, the present-day distribution ‘of primitive genera in a given family or even primitive species in a given genus is of as much, or more, importance as the present-day distribution of primitive iamilies. We unfortunately lack a sufficient knowledge of the compara- PNAS aicrt ii Re Sica maw iS * 7 RE TNE = es Bey ees OS is ll R ee CAL Me] Ae Fic. 5.—Distribution ef Archeide A, Fossil (amber) Archea; B, Living Archea; C, Mecysmauchenius. tive anatomy of spiders to enable us to use such data in the solution of the problems of distribution. In fact, widely discontinuous distribution is, at present, the best indication we have of relative antiquity and its use is rendered hazardous by reason of the possibility of polyphyletie origins—a possibility which has been rendered more probable by recent work in experimental evolution. An analysis of the distribution of the world’s spider fauna would be out of place in this paper and only a few points which seem to have a bearing on the West Indian problem will be taken up. The family Archeide is an interesting one in this connection (see figure 5) since Baltic amber contains a genus (Archea) of which the LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 13 2 only known living representatives are in Madagascar. Furthermore, the only other genus of the family is confined to Patagonia. Had it not been for the fortunate find in Baltic amber we would not know that the family ever occurred in the northern hemisphere and a land bridge from Patagonia to Madagascar in the most direct way compatible with sea- bottom contours would receive additional strong support. As it is, it seems more probable that this was once a fairly widespread family and it certainly occurred in northern Europe. We know that prior to the glaciers the climate of the North Polar regions was mild or even tropical Fic. 6.—Distribution of certain genera known from Baltic amber 1. Segestria; 2, Dysdera; 3, Eresus; 4, Amaurobius; 5, Archaea; 6, Agelena; 7, Anyphena. Only certain points in some of the ranges are indicated. The lines represent possible routes of dispersal. and hundreds of circumpolar species point to the interchange of fauna between Eurasia and North America, where land bridges, if they ever existed, would not need to be long, so that it really seems quite probable that the genera of Archzide which are found in Madagascar and Pata- gonia respectively are merely remnants either of a formerly widespread family or of a northern family which was driven south by competition with new forms. Such a movement is supported by mammalian fossils and seems more likely than a South Polar connection. Reference has been made to Pocock’s list of seventeen recent genera found in Baltic amber. Of these we cannot consider here Aranea and 138 ANNALS NEW YORK ACADEMY OF SCIENCES Zilla because their taxonomy is too unsatisfactory. Drassus has been split up by authors but even one of its parts is practically cosmopolitan. Tetragnatha, Tegenaria and Philodromus are cosmopolitan, even includ- ing frigid regions. Olubiona is found throughout most of the temperate and tropical regions ; Nephila and Sprassus are cosmotropical; and Thom- isus is found throughout most of the continental, at least, Old World. Certain points in the ranges of the other seven genera are shown in figure G, only those most distant from the Baltic together with several inter- mediate stations being indicated. The supposition that all these seven Pcs Paes eRe Sei sea LCC eR Hint asc ot eee la : ee S 5 z 2) | yy { e | Sy G 1 ES Hic. 7.—Distribution of Segestriine Lines refer to Segestria and dots to Ariadna. cenera originated in the Baltic region in, say, the Eocene would be ab- surd, but supposing that they did, would not the several million years since then, including as they do a long space of mild or tropical condi- tions in the Arctic, have been sufficient for the dispersal of these genera more or less along the lines indicated in the figure? I feel that even the rather well authenticated land connections in the North Polar region would not be necessary, for we must remember that the distances are much shorter in the polar regions than they appear to be on a Mercator’s projection. Certainly, there seems to be no need for Antarctic connec- tions. The distribution of Archea has already been mentioned. The only other genus of special note is Segestria. The discontinuity of its LUTZ, LIST OF GREATER ANTILLEAN SPIDERS 139 distribution is toned down by a consideration of Ariadna, a genus which with it forms the subfamily Segestriine. Thus Segestria in New Zealand is linked with Asia by Ariadna in Australia, Sumatra and Ceylon. See figure 7 By selecting certain genera or groups of genera of spiders it would be easy to make quite a list in favor of a land bridge from South America to Africa and elsewhere such as that shown in figure 8 which is after A. E. Ortmann’s idea of certain of the connections in Upper Cretaceous times. For example the following genera might be cited: Cyatholipus 2.02 ECON TESST RNS! TV DT 0 C2 a (0210 0 ne 1.46 Averace Shell mdex (adult) occu: c.sasswemweves (about).. 1.80 BaP west SHEL Index CHEAMIC StHHOE) <6... secre eens sewn ns 3.50 lowest shell index (meanic stage)... 05. ccc ce ke ee ek 1.78 Average shell index (neanic stage)............ (about).. 2.65 ANNALS NEW YORK ACADEMY OF SCIENCES 8 el IsuaUad|yD “~nue SNyDUOLONU AafLMdy JO sabnD)S d1UDAU PUD ZINpY UL SadipUL Jays JO suOYIPWOD ArtzDADAWOD BuMoYs saaingQ—T ‘DIyy a {, 4 SIZF—-/0F ‘00S-94T SLE—-LE oS) 2 : 00t—9L/ 294] —EF/ S cS I —~ 8 ~~ ~™ ( ~ S 3 a] ~ MOOK, VARIATION IN SPIRIFER MUCRONATUS 179 TABLE I1.—Number of individuals and percentages of total number of individuals of various indices : Adult Neanic Index a Number Percentage Number Percentage N22) ed L235 | Doe ee mn eee 3 16.3 0 0 eM FN 2a hatenater io 2 bor rae Fic. 2.—Curves showing comparative conditions of shell indices in adult and neanic Stages of Spirifer mucronatus mut. multiplicatus Mutation II1.—Spiifer mucronatus Mut. profundus GRABAU Ms. This mutation occurs with attenuatus in the upper shales in the Mich- igan region. It is small with extremities not at all or only slightly mucronate. The median sinus is deep and angular. It extends from the beak to the anterior end, then beyond into the brachial valve. The bottom of the sinus is subangular. The fold on the brachial valve is shallow and has no median depression. The plications are round and close together. There are 8 to 12 on each side of the fold or sinus, 180 ANNALS NEW YORK ACADEMY OF SCIENCES rarely less. The growth lines are strongly turned back over the fold and sinus. Mehr a yh Elsner ree ene ee te aa 35mm. Miniimiim+ awad tins tec exesee ese re ene ee tee ee eee eee Sieue 15 Maximum Wemerthics eke ers send ees a cai eae retene 21 Minimum feneth:3 3 che ao ies cee ae eee eee 9 Highest: Shell index -(adult) fancies oe nee one eee eee Pape bowest shellindéxs (duit) 325. eee ee eee eee 97 Average shell index (adult)......... eee eee (about)... 1.65 Hichest shell index, (neanie stage): .-2..cen = are eee ae 4.37 iowest shell index (meamic staze).-2.. 4-400 cee eee 1.35 Average Shell index (neanic stage)............ (about)... 2.50 The adult curve is fairly short and regular. It corresponds closely with the adult curve of alpenense, though the average index is a little lower. The neanic curve is very broad and irregular. It corresponds rather closely with the neanic curve of alpenense. ‘The average neanic index, however, is a little lower. Two hundred and seventy-six indi- viduals were measured. TABLE III.—Number of individuals and percentages of total number of individuals of various indices Adult Neanic Index (ea ee — Number Percentage Number Percentage SOO. aie S sapere nes Seoeeem enters 2 7 0 0 } OASIS ZB, recite tk ere ec a ager 36 13.0 0 0 1 ZO Ad DO ee ah ee eee are ee eee eas 112 40.5 2 Bart he aera as Say seek eae eae: 75 Zt Al 6 pa | LA O22 B00 se sidtovaee ie ere Cerne aici 40 14.4 37 13.4 Pr ONE De siar opeeta oa ea eee ste eee teers 6 rea | 72 26.0 Di PAGE OO ors Ek SRE te eating Sete 3 1.0 52 18.8 Dis By NEDA < velianeiscusvsanon Ovepeaseaae se eeunercgane ie 2 at, 60 21.5 2 ABS LOO! s UPS, GeO Reeves arene eases 0 0 31 11.2 3 OISS 22) ceo oe ee ee eae 0 ) 8 2.8 DOS OO eeetesanien svete aera Ss euaegeate eee 0 0 = 1.4 So Chl es ALO ai austa Cle rord esters ate es tate eee see 0 0 0 0 BOAO aoe nein eee oe ee eee 0 0 2 eit A ON AAD hig Ste ilo whe Serene Cee ues 0 0 il. 3D 42GB SOF ogc alk aie whe oA 0 0 1 .o0 Morals. Wurcs: canieeeas is ead oe 276 99.5 276 99.3 This mutation is evidently an offshoot from the alpenense stem. The line of development has been more in the nature of a deepening of the sinus and strengthening of the growth lines rather than in reducing the shell index. The shell index is a little lower than that of alpenense in 181 MOOK, VARIATION IN SPIRIFER MUCRONATUS 7 snpunfoud “Qmik SsnaDWOIONU MaJIadgy JO SaHN]8 OUDIW PUD WINpYD We sadrrpU Yays Jo suo IpUOD dAYDADdWMOod OWMOYs Sari O—"e “DIT + w w cA w is) bi w re) =< a a 2x4 be g R 8 g x a ; x a & s t ' 1 | 1 t 1 ‘ J ' | - - - w w w +) LS) b b = =~ — haere : : 5 : : 5 : | : : 3 ee 2 a i: o- a ai hg ; “bt 182 ANNALS NEW YORK ACADEMY OF SCIENCES the adult stage, but only a mere trifle lower in the neanic stage. This in itself is an indication of progression toward a low indexed condition. Mouration 1V.—Spirifer mucronatus Mut. thedfordense SHIMER AND GRABAU The mutation thedfordense occurs in the upper Hamilton shales im the region about Thedford, Ontario. This mutation is the most advanced of the five mutations studied. It is fairly good-sized and very stout. There are 12 to 15 strong sharp plications on each side of the fold and smus. The sinus is deep and subangular in the adult and occasionally has a faint elevation in the very young stages. The fold is somewhat elevated and rounded in the adult, and often shows a faint median depression in the young stages. The growth lines are very prominent, and are sharply deflected over the fold, sinus, and plications. Two hundred and thirty-one individuals Were measured. Maxinigm: wiGth. ...'3 <25< 2 si. oo eee eh ee ee eee 45mm Maximum Tengtlt oo 2253 pos See eee ae ee eee 35D Minimum feneth) . 22; 22 ee ween ee eee eee eee eee 15 Highest shell index: (adult) co2.. = ae oe ee ca ene eee 1.94 Lowest shell“index’ (adult). 2. fe2n se ae oe cl eee 89 Average shell index ¢adait) . veo 26 54-0 eee te (about).. 1.30 Highest shell index (neanic stage)...........-.........-- aoe Lowest shell index. (meanic stage)... 06.2... . .s'-. « 22 eee 1.24 Average shell index (neanic stage)............ (about).. 2.00 Taste 1V.— Number of individuals and percentages of total number of individuals of various indices Adult Neanic Index (pee Number Percentage Number Percentag = OO oo ob. Sis oe ee eres 24 10.3 0 0 AE 2S fw oe We ee ee ee 107 46.3 2 Cae: 2 2D BD). 3 ao oaths bine eee ees 80 34.6 17 tock Be TE oii See oe ee ee eee 15 6.4 54 23.3 PEGS OD oh seed Be eee ee 5 2.1 63 26.8 DE os Seiden = on eee ee 0 0 40 17.3 2 BD nt & ues ee eae eee 0 0 24 10.3 RD) SiS hae Dine Sore ew See ee ee 0 0 18 t.8 2 FGA so ok es Jae eee ee 0 0 10 4.3 SS OES 3S 8 kd etek tes eee ee 0 0 -: 4 5 ap, ae ae | eee ee Sees We LPT S 0 0 1 .4 Cet ee Sy ee eer ee a 0 0 1 4 231 bac] o ct DN) — i) Ww pot =| OSD SPIRIFER MUCRONATUS MOOK, VARIATION IN asuapso{payy “gnu snqouowonm safindgy fo sabn7s ovunau pun 4yjnpy UW saavpuy qjays fo suo1Ipued daYnundwmo0d HBuMNoYys SAA~AND—F “DIT OSE-FTE *Hesee -/08 '00E—-ILZT q N x ; : S2r-loet 000-941 3 y 0.$[-Fe/ -00/- 184 ANNALS NEW YORK ACADEMY OF SCIENCES The adult curve is very narrow considering the large number of individuals measured. It is regular in form. The neanic curve is fairly regular and is very wide. It would correspond fairly well with a flat- tened-out curve of the adult. Thedfordense was evidently derived from alpenense along with pro- jundus. It differs from profundus in having vestiges of a plication in the sinus and groove in the fold, in being more robust, and in having a lower shell index. Profundus has progressed further in the deepening of the fold and sinus, and thedfordense further in reduction of the shell index. Mutation V.—Spirifer mucronatus MUT. attenuatus GRABAU MS. Attenuatus occurs in the upper shales in the Michigan region along with profundus. The shell is very wide and short, giving a high shell index. The sinus of the pedicle valve has a distinct median plication which extends nearly to the beak. The fold of the brachial valve is flat and level with the plications on either side or slightly elevated above them. It has a pronounced median groove which extends nearly to the beak. The growth lines are moderately strong. There are 15 to 18 strong plications on each side of the fold and sinus. Maximam-wiith.: 5 sé. ..0. tee as ees hee oe eee eee eee 47mm. Minimo width: o s/s 23 SEC 2 ation See ee ee eee eee 19 Masimum Jenpiiie <9. 2 ¢eet Soe eee Ce ee oe eee eee 17.5 Minimum Jeneih: 2 pods. eee See ae eee bee eee 8 Highest shell index (adult) 22:02. S.. c.ce2s Jes sector 3.75 Lowest .shell index \(adett) 2202.75. ee ees ou. ee eee ay) Averace shell index -(adult) a2 02. eae eee (about)... 2.25 Highest shell index ‘(neanie stage)... {2622 25.22. sec Seca 4.70 Lowest shell index (neanic Stage). ..252-...2.s-c..eeeee yA Average shell index (meanic stage)............ (about)... 3.40 TABLE V.—Number of individuals and percentages of total number of individuals of various indices Adult Neanic Index ed Number Percentage Number Percentage 1 Pe ce od Sk so ee eee 2 2.0 0 0 6 ed ni en cele cee eee 7 9.4 0 0 De oe Recs ee aie eneiie me 16 21.6 0 0 2 2 Rs ak Swe we ok eh oe eekoae 1s 24.3 1 1.3 2. ne ome See ow ae eee oe 14 18.9 6 8.1 2.2 TE: code oe ates eee teens 10 13.5 7 9.4 2 0D. ie oct as vs kee een eee - 5.4 13 17.6 ay oo a Be do ds ay ce) bape te) “1 i Ht © bo veo) 185 MOOK, VARI y ATION IN SPIRIFER MUCRONATUS SNYDNUAIAD * 1p “Qn Sn.DWO wonw dafiud wdg jo sabn 18 OVWwDAW | 1 pun .)npD UW SadUpuUr 1 Q yur rays jo s WOVUApUoOd a: yi “UzwIvawUo0d b 1 9 Buaror 1 {S SOLMMQ— VAN Q—G “OTH a ra C+ ae eee : . E ‘ 3 by q : s 8 : N : 1 7 i v Fr x bs ® 3 x 5 : i : “yr : 4 J Ne & Ve 1 < x a e a ——— e s x S l ri q & AA Ree: q Non CDAD RE eee ° u S : a : : ae 7. Lv 3 : 9 Ny Z . Ligh | A 3 ae +s. ae “ : 3 a 7 c* 2 79 3 i) aia | iD at hey aan eo V8 =< ta po Vel ara ue , ate ale r a le sets . m nae Fae ioe ae AS bet eras cs cay ‘ aS ‘“ i age coe Et 186 ANNALS NEW YORK ACADEMY OF SCIENCES Adult Neanic Index ee Number Percentage Number Percentage 3 .b1-S. se coe oe ee eS ee 1 1.8 10 13.5 31642 00 25 eee ee oe Eee 0 0 ) 6.7 4 O14 2S ih ees He ee SE 0 0 2 raf B24 BD Ss ae Se Eee See 0 0 » 6.7 4c Dl (De oee SSEe Soeeeeeee 0 0 2 2:7 otal on ces ote eee ee T4 99.8 74 99.7 The adult curve of attenuatus is broad and rather irregular. The neanic curve is very broad and irregular. This mutation is in many respects the most primitive of the five muta- tions studied, although it occurs in the highest of the three horizons. It is nearest to multiplicatus in form and general characters though multi- plicatus has a lower shell index. The derivation of attenuatus is uncertain, though it is not far removed from multiplicatus. Several possibilities may be considered. Attenuatus may be derived from some form of muj/tiplicatus whose index is much higher than any of those measured. Jt may have come from a multi- plicatus stock having an index like the highest of those measured and then degenerated into a short and broad form. It may have sprung from a common ancestor with multiplicatus in a lower horizon and re- mained primitive while mwultiplicatus progressed toward a lower index. The third possibility is by far the most probable. The adult curve of attenuatus corresponds roughly with the neanic curve of multiplicatus. CoNCLUSION Two distinct lines of development have been observed in Spirifer mucronatus. In one of these lines development has advanced to a con- siderable distance from the primitive Spirifer mucronatus condition. This line includes alpenense, profundus, and thedfordense. The other line has progressed much more slowly. This line contains multiplicatus and attenuatus. In the first line thedfordense and profundus have prob- ably been derived directly from al/penense. In the second line attenuatus was probably not derived directly from multiplicatus, but from a form closely resembling mu/tiplicatus, but more primitive. Multiplicatus and alpenense were probably derived from some primitive form of Spirtfer mucronatus belonging to a lower horizon than any of the five mutations studied. The approximate relations of the five mutations to each other are shown graphically in the following diagram. MOOK, VARIATION IN SPIRIFER MUCRONATUS 187 attenuatus profundus thedfordense multiplicatus alpenense Primitive form of Spirifer mucronatus Fic. 6—Diagram showing the relationships of the five mutations of Spirifer mucronatus The tendency in development has been to reduce the shell index, to reduce the number of plications,-to lose the groove and plication on fold and sinus, to strengthen the growth lines and individual plications, to deepen the sinus, and to a certain extent to reduce the actual width of the shell. ANNALS NEW YORK ACADEMY OF SCIENCES 188 Snynoydyinm ,, 9 99 PSE) O10 ase 8 0) a snpunjoud “yn snpouosonm safimidgy Jo suoyngnu auf fo sabnys ynpp ay. Jo w ics} & A : 4 g = iS 4 R [ ! \ cy) Gy A) N 3 : iy 8 5 3 HON a 5 Cael} 5] AiNiainietn ar raiione Gen AR Le * ” —- — — — — asuauadyo ,, 99 99 ‘xopul yjupe = ———_———____ 9suapuofpay, “ynm saoipur ays JO suonipuod —daynsndwmo0d HmMnoys saavgo— ), 00% - 9L/ 97 (-/S/ S7/-10/ ool ‘DIA 189 SPIRIFER MUCRONATUS MOOK, VARIATION IN ” ” RA RA ARARARA 8NYDNUAIIO ” ij a) 0: eee ee — snqvoydyinw ,, vs x =o asuauadjo ,, ‘SOOUP OP MUO) cease naeeseumeeces snpunfowd "yn ‘xopuy syuvou =——_——____—___ asuap.sofpay? *ynu1 snpnuowonu Jalrudgy fo suoynenw auf lo sabnys aunau UW saoipur ays fo suo1Nypuod aaynsndmos HBumoys saainp—s “DI S & 2 ! cr el “00 f-9AT YG ded AY ‘002 -9L| StE-—10e ‘Stl -}9/ 190 eeeee see ee eceeccece eeceee or) Co Co CO ~l ea) ede oeeee eeeee ceaeee «eeee oeeee ©“eeee “eeee ANNALS NEW YORK ACADEMY OF SCIENCES ‘TABLES OF MEASUREMENTS AND INDICES Npirifer mucronatus mut. thedfordense Adult Stage Neanic Stage (Es si =e aN (Ge Width Height Index Index Width Height Index Index (reverse) (reverse) Simm, 929mm 72935 iL. O06 Simm, 23mme Sc: 1.34 32° 29.5 921 1.08 32 Z0es .640 1.56 Bye ie .843 LS 32 AT, Sea 1.88 38.9 27 Sabi 1.42 38.5 20.5 02 TSG 30 3 1.000 1.00 3 24 . 800 1.25 29 ZTE) 948 1.05 29 eas O72 1.48 31 26.5 bd 1.16 31 19 612 1.63 3+ 28 .823 te, 2. 54 23.9 .691 1.44 om Bee 1.015 .98 32 18 . 962 WAAAY 2 2B Bs ail! lial QT die .629 1.58 54 24 TOD 1.41 54 a9 OOS 1.78 39 530.5 871 1.14 39 19 542 1.84 39 28 . S00 1 WBE) 35 Ay, ey .500 2.00 38 24-5 . 065 1.76 38 16.5 434 2.30 5- Pit . T94 ha2b 34 Ze 647 1.54 30 26 S66 JERR) 30 20 . 666 1.50 Ste (e539 723 8's) 38 18 473 pipes 1 32 28 S75 1.14 on 20 .625 1.60 29 De 793 126 29 1935 “G2 1.48 33 2ORS 893 alee slat 33 48.5 .560 1.78 29 24 827 1.20 29 IPT Ges) .603 1.65 30) 25 .t14 1.40 oD 18 514 1.94 36 3 S61 1.16 36 aaah .583 ge (a. 28 24.5 S75 1.14 28 17 . 607 1.64 ita) 26 742 1.34 35 Pepe .628 1.59 32 23-5 T3834 Weals Be a L7G m5 )3) | 1.88 Pat 2 851 2 ae of 19 . 7038 1.42 29 aH | 931 1507 29 Di . 124 1.38 40.5 28 .691 1.44 40.5 ZO KES 506 1.92 Beas, 3 . 923 1.08 ao Je a Wega: 038 1.85 28 19 678 1.47 28 15 585 1.86 a1.5 Zed 904 1.10 3) Na) 18 ay ial ies 38 50 7TS9 1.26 38 23 . 605 1.65 26 20 740 tao AT eas 14 .518 1.92 35 23 657 1. 52 50 18 514 1.94 31 20 645 dea ak 14 451 at I Ot se 26 693 1.44 S120 20 N30 1.87 40 32 . 800 120 40 2a 575 Lis mk A) * 806 1.24 ou VE 548 1.82 a2 y) 906 1.10 32 19 .993 1.68 Number Peeves 69 aS eS es es S| He wwe =] Or eeceese eeeee eeeee MOOK, VARIATION IN SPIRIFER MUCRONATUS Width Spirifer mucronatus mut. thedfordense—Continued Ol Adult Stage — Gn, SE Ee aE Index Index Height (reverse) 26mm. _ .896 pa Ba: 3 . 967 1.03 29 1.054 .94 25 .781 1.28 26 . 896 tM 25 hee 1.32 24 . 750 1.33 25 . 833 ¥.20 22 745 1.34 26 702 1.42 ra | .870 1.14 24.5 . 662 1.51 24 . 800 1.25 25 tSk 1.28 21 750 1.33 23.5 734 1.36 26 787 1.26 23 . 851 eke 26.5 . 883 iis 26 . 962 1.03 23 .676 1.47 25 .625 1.60 24 Ae of 1.37 30 . 769 1.30 24.5 . 844 1.18 21.5 .53T7 1.86 29 . 125 £-3¢ 27 .710 1.40 21 . 750 1.33 26 . 742 1.34 25 .641 1.56 29.5 . 867 1.15 hs ae 1.23 25.5 . 980 5 AT 24 .8138 1.22 25 . 862 1.16 25.5 . 879 1.13 25 . 806 1.24 16 ool 1.81 15 . 930 1.86 21 . 646 1.54 25 . 806 1.24 ° 20 -588 1.70 27 Height Index (reverse) 20.5mm..706 PA! 677 20 Be fs 18 . 962 17 .586 Ve 015 19 .593 18 . 600 14 474 18 .432 20 . 645 15.5 .418 19 . 633 15.5 .484 16 .OT1 18 .562 19 075 15 .055 17 .566 16 .592 18 .529 19 -475 7 515 20 512 18.5 .638 LG25 437 24 . 600 20 my AS a rg . 607 LFS .500 18.5 474 21 617 20 . 625 15 .576 13.5 457 SES5 . 603 15.5 .596 i .548 12 .413 oF -392 17.5 .538 18.5. .G29 15 441 Neanic Stage Se 191 Index is NO ce oN Oe GO el el 0 oll oO OSS “I-91 WwW Ap Soa a al 192 ANNALS NEW YORK ACADEMY OF SCIENCES Spirifer mucronatus mut. thedfordense—Continued Adult Stage Neanic Stage — aay SS Number Width Height Index Index Width Height Index Index (reverse) (reverse) ‘oie Iperenieee 33.5mm.26.5mm. .791 1.26 33.5mm.20mm, .597 1.67 SO eS 28 PALA) acre 1.30 28 17 . 607 1.64 GO keer 30 30 1.000 1.00 30) 19 . 633 1 52 os ae 34 25 735 1.36 34 18 .529 1.88 Se 39 27 . 692 1.44 39 19 .482 2.05 ORs cris 34 26 . 764 ool 34 19 .5d8 a eT fe: oe age 34 25 pasa) 1.36 34 19 .558 148 Oy Le ote 34 23 .676 1.47 34 14 .411 2.42 SG ieee 43 29 . 674 1.48 45 19 441 2.26 a earerge ol 30 . 967 1.03 31 18.5 .596 1567 Seo ees ao 24 .685 1.45 35 aS .500 2.00 OO ere isie DAE 26.5 .981 1.01 2 19.5 tae 1.38 HOO S. se5 39.5 Bye) . 822 gyal | ae 19 O84 ae: 101-175 inclusive — Spirifer mucronatus mut. alpenense. PAS Mae 33.5 32 952 1.04 33.5 20 597 1.675 P| V2 eae 20 Dianed . 944 1.05 2a 14 .518 1.92 DM £5 Fee ae 33 . 968 1.03 32 19 .593 1.68 AG fe ae 26 28 1.076 .928 26 18 . 692 1.44 ZA We Se? 30.5 28 .918 1.08 30.5 14 .459 2. Mik BES S08! 33.5 32:5 .970 1.03 33.5 20 .D9T 1.675 51 eee 26 26.5 1.019 .981 26 14.5 Dot 1.49 yd 1 eA 35 33 .942 1.06 35 22.5 .642 1.55 pit 1 RN Laas Zt Zit 1.000 1.000 27 18.5 .685 1.45 DEAS borows Si 29 - 985 1.06 31 20 . 645 1.55 PaaS Brae 26 2 . 961 1.04 26 17.5 . 673 1.48 De er sats 28 26 .928 1.07 28 2st) . 803 1.24 Fy hove eis 29 26 . 896 Kolt 29 20 . 689 1.45 PAG gees By 31 . 968 1.03 oe 22 .689 1.45 227-445 inclusive — Spirifer mucronatus mut. alpenense. 446..... 39.5 26.5 .670 1.49 39.5 iy G5) .4438 2.25 447..... 3 pape . 666 g Meta f axed 14.5 .439 pee F SAS es. 35 ok .885 q oe 35> 17 .485 2.05 449..... 36 291.5) . 708 1.41 36 20 .555 1.80 ABO te sss 36 22 .625 1.60 36 16.5 .458 2.18 7 15 Ore esis qe ye ee echt heme «tates esta 4B nos a 33 23.5 th 1.40 33 16 484 2.06 ADS ania a's 37 28:0 770 aL 29 37 18.5 .500 2.00 454..... 30 24 . 800 L.25 30 17.5 -583 oe fs ADD os os. 34 27 . 194 25 34 22, 647 é Aa: co Oe 36 AL a . 763 1.30 36 20 556 1.80 AML 5.0 65 32 221.9 . 703 1.42 32 14.5 .453 2.20 BS ot anccs 35 27 errs 129 35 19 .542 1.84 4D oc0 5 31 25 . 806 1.24 31 18.5 .596 1.67 AGO. PS ci 45 30 . 666 1.50 45 21 . 466 2.14 MOOK, VARIATION IN SPIRIFER MUCRONATUS 193 Spirifer mucronatus mut. thedfordense—Continued Adult Stage Neanie Stage | aa -A~~ a FSA, =e >. 29 17 986 1.70 29 10 344 Gik 2 oe 32 35) 1.093 914 32 15 468 G2. oS 29.5 26:5 S9S teat 29.5 14.5 491 5 eee ae 354 1.096 wi! i Bf 31 Hy .532 le eee 38 31 $15 122 38 13 342 Geet cs 31 25 S06 1.24 31 255 403 Gabi 25 ys 1.040 .961 25 11 440 GaSe 2.x. 30 31.5 1.050 952 30 16 503 it) eee 24 20 33 1.20 24 9 379 1 sae rt | Zico ss 981 27 12 ait te. 37 19 514 1.94 37 11 297 GS...) = 27 S70 1.14 3 12 387 664..... a Giese cs 30.5 20.5 672 1.48 30.5 13.5 442 666..... mpegs of? ve F TCT ire 30 29 . 966 1.03 30 15.5 .516 Pee 24.5 A oh). ASSES .89 24.5 14.5 . 604 7 TS See 27.5 27 .981 1.01 21.5 16 581 Gre. = 1G: rf 93 1.14 = a 16.5 523 ye im hae 30 2L.5 716 1.39 30 15.5 516 Gia Ss S41 yf (ee SST 1212 si. 14.5 467 Eee 40.5 27 666 1 50 40.5 16 395 ifs ete 31 24.5 790 1.26 31 17 548 1s eegeaione 3 24 .610 1.62 Sy 16.5 423 Giles: 35.5 3 .§45 1.18 35.5 17.5 .494 yt Rae oe 3.5 27 .620 1.61 43.5 17.5 .402 ire. ox cae 26.5 736 1.20 32 EGS 343 yee 31 23 .D 754 1-31 31 1 .419 ic) ee 39 28.5 i3 cE 39 16.5 428 4 eee 37 25.5 .689 1.45 37 14 378 6S2 40 28.5 712 1.40 40 13.5 437 Git 2% 34.5 25 724 1.38 34.5 14 405 684..... 34.5 ry 197 1.25 34.5 15.5 .449 27 ee ae 37 29 i83 i 37 16.5 445 ites =: ae ot 25.5 689 1.45 37 14 378 oy aa 32 32 1.000 1.00 32 18.5 578 12 ae 38 26.5 .697 1.48 38 13.5 .355 2 ee 31 28.5 919 1.08 ol 17.5 - Dae ci | iti 54 23.5 691 1.44 34 14 411 MOOK, VARIATION IN SPIRIFER MUCRONATUS 195 Spirifer mucronatus mut. thedfordense—Continued pdult Stage Neanic Stage Number Width Height Index Index Width Height Index Index (reverse) (reverse) Coe eee 36mm. 2imm. .7T77 a Ae: 36mm. 15.5mm..43 2.32 GO2e 6 29 26 . 895 cee 1 29 14.5 500 2.00 BOG saws Sa <0 Diy .T61 io ao 12.5 “ote 2.68 OS See 28 pee | . 964 4303 28 14 .500 2.00 CS ae 35 29.5 .842 #48 35 af .314 SAT BGS. isis < 31 22, . 709 1.40 31 i 5 skO 2.69 Co ee 37 22.5 .608 1.64 37 t225 .337 2.96 ols eae hes Seis ee Sey ate te tS ieee 39 26.5 .679 1.47 39 13 .330 3.00 0 er 30 30.5 | 7 1.016 . 983 30 15 .500 2.00 oe 30 23 . 766 1.30 30 13.5 .450 Zee le 42 34 . 809 2S 42 15 soot 2.80 Rates's s ee BS os ote ee A eee aenke wet Seah Wate... 36 20 . 805 1.24 36 13.5 mY (3) 2.66 (A ee 31 aL.5 1.016 . 984 31 14.5 . 4038 2.43 7.0 31 29 .935 1.06 31 155 .435 2.29 7) ae ee sicher =e ses Bes: eae d DIOS 6 ose = 35 23.5 Boy! 1.48 35 ib 428 2.0 “(tS 34.5 28 .811 1.23 34.5 18 O21 197 iC) an 31.5 26.5 . 841 1.18 31.5 15.5 .492 203 ti 30 24 . 800 Rees 30 ee .416 2.40 oe 36 so.5 1.0138 . 986 36 18 .500 2.00 WAS es 3 ss 36 29.5 .819 122 36 16 .444 2.25 ee... 33.5 29.5 . 880 1.38 33.5 15 447 2 De Aas. Ss 29 26.5 913 1.09 29 15.5 536 1.87 Gos: . 33 29 878 iets 33 14 424 2 3 ‘i a 36 30 833 1.20 36 13 361 2.76 “ae 40 ry es) 687 1.46 40 15.5 387 2.58 Ge. os 41 29 (07 1.41 41 TL5 .280 3.56 (a 38 29 763 fi 23 38 14 .368 2; Th i) Sa 41 29 .3 714 38 41 16 .390 2.56 ees. Sor bots Sahreln ehectvs tyne ogee Mea ss 29 25 862 1.16 29 13 .448 2235 Spirifer mucronatus mut. alpenense _. ee 38 22 918 £272 38 14 oOr pApsy fi | : 34.5 20.5 578 72 34.5 125 . 362 2.76 ee... 35.5 19.5 . 549 1.82 35.5 13 . 366 28 mete. Ss 34 A 53 easy 34 12 -nehen 283 Re ss. 32 pa a] 640 156 32, 14 437 2.28 2. 29 17 586 5 ar 29 12.5 .431 2232 ees... 38 1-5 D138 1.94 38 12.5 .328 3.04 Lae 27 16 592 LS AT 11 . 4038 2.45 i ae 32 23 718 1.39 32 15 .468 24428 196 Number 3 = (sei te “eevee *| = 6) os "ee ee CPeevoevee “ee ee = se ee w a terel es ae a ew Width 29 OT Adult Stage a Height Index Index (reverse) 22MM, iso 1.36 20.5 . 500 2.00 did oe .442 2°26 20.5 . 640 1.56 17, . 680 1.47 18 .620 1.61 19 642 f 263 a eS ool d voy s 18.5 .486 2.05 Be . 647 1.54 22 .68T 1.45 24 . 600 1.66 i) .500* 2.00 22 .o94 1.68 20 .506 LIF Zs . 833 1720 23 ese 1.34 21 617 1.65 23 657 1.52 18 . 692 1.44 ps 3) Bu, 514 1230 23 - 93 1.26 22 785 127: 13 . 464 2.15 23 . 884 TAS i | . 700 1.42 21 677 1.47 195 973 1.74 25 961 -1.04 23 . 605 1.65 23 . 766 1.30 20 . 689 1.45 25 .TS1 1.28 22 .814 ti 22 21 . 636 1.57 16.5 932 1.87 22.5 . 803 1.24 26 .812 1.238 20 . 666 1.50 13.5 540 1.85 20 . 83: 1.20 a . 124 1.38 12 .415 2 deh ‘Width 30mm. ANNALS NEW YORK ACADEMY OF SCIENCES Npirifer mucronatus mut. alpenense—Continued Neanic Stage os Height Index (reverse) 15.5mm..516 12 — 15 a i Ol OT OT OF OT OT . 292 . 288 .468 . 460 344 387 084 ew ww. Ww: OR Ne Od Se) NOWwWnWwnwnnwr Ww wwnpew Www ble bo Nw eH & wNomwWM Wh WY Ww bw Wwe bd be MOOK, VARIATION IN SPIRIFER MUCRONATUS 197 Spirifer mucronatus mut. alpenense—Continued Adult Stage Neanic Stage Width Hane Index Index Width Height . Index Index (reverse) (reverse) 30mm. 22mm. .733 1.36 30mm. 14.5mm..483 2.06 54 3 -382 2i Gl 34 S .235 ote 29 15 ea G 1.938 P49) el? 318 2.63 26 14.5 ont 1.79 26 10.5 p11 2.47 oe 16.5 912 1.938 32 12 oto 2.66 34 18 -029 1.88 34 12 .352 2.83 30.5 23 .613 1.62 37.5 13 . 346 2.88 34 18 .529 1.88 34 10.5 .388 3.23 26 15 016 1.73 26 8.5 .326 3.05 29.5 17 576 1.73 29.5 10.5 3 193) 2.80 3 19.5 629 1.58 31 IZ .5 .403 2.48 on 22 687 1.45 32 13 .406 2.46 28 26 928 AO7 28 13.5 482 2.07 26 18.5 711 1.40 26 10.5 . 403 2.48 32 27 . 845 1.18 32 16 .500 2.00 26 23.5 . 903 1.10 26 AT . 653 1.52 15 9 . 600 1.66 15 ic .466 2.14 29 17 .586 1.70 29 cia .379 2.63 30 18 600 1.66 30 ala! . 366 2.2 ok 20 645 1.55 31 15.5 .500 2.00 36 18.5 5138 1.94 36 12.5 .346 2.08 29 18 620 1.61 29 11.5 .396 2.52 37.5 22 O86 Bef 37.5 11.5 .306 3.26 23 24 1.048 -96 23 14.5 . 630 1.58 yy 20 909 1.10 22 B35 .522 1.91 39 23 589 1.69 39 16 410 2.43 39.5 16 405 2.46 39.5 11-5 291 3.43 29 22, 758 gtr 6 20 16 551 gE | 32 19.5 609 1.64 32 13 406 2.46 38 15.5 407 2.45 38 ae 289 3.45 30 3 .433 2.30 30 9.5 316 3.15 or 18.5 596 1.67 oo 14 451 2 36 20.5 569 1.75 36 138 361 2.76 28 19 678 1.47 28 13 474 2.15 36 18.5 513 1.94 28 on5 410 2.438 32 21 656 1.52 32 14 437 2.28 28 15 .5385 1.86 28 12 .428 2.38 28 22 785 1.22 28 9 321 3.11 ah) fan) jt eo) OO ey | bom bak yea ow fan) feed OT bm bed =r) ho bem fom) ses ee = P'S ins "eee "see *= es @es Adult Stage Width ON Height Index Index (reverse) 16mm. _ .533 1.87 28.5 .950 1.05 20.5 .500 2.00 20 645 1.55 20 645 1.55 19.5 .513 1.94 17 43 2.05 21 538 1.85 24 750 1.33 21 717 1235 20 588 1.70 21 .750 1.38 28 .875 1.14 22.5 sis i 19.5 .750 Be= Ws 653 1.52 17.5 .480 2.17 23 621 1.60 19.5 .590 1.69 24 827 1.20 20.5 .525 130 22 . 709 1.40 raysss) 1.80 19 .452 2.21 25 .851 i Bea Wg 18 . 600 1.66 21 . 700 1.42 20.5 .625 1.75 19 .465 2.15 Width 30mm. 50 41 31 ol i) 2 dS Ww oo ip) i) =r) “I ND 0 Or ANNALS NEW YORK ACADEMY OF SCIENCES Spirifer mucronatus mut. alpenense—Continued > Neanic Stage Height 13mm. cL 16 14. 14 a or Ot 19) | Or Ot Index (reverse) .433 .516 .990 . 467 .451 .328 271 294 NW po bo = Oo me WwW OD 9 9 Br a! oH 9 19 eo 09 NY ty BSE hm OD OO be mono SSK MOOK, VARIATION IN SPIRIFER MUCRONATUS 199 Spirifer mucronatus mut. alpenense—Continued : Adult Stage Neanic Stage Number ‘Width Height Index Index Width Bene “Index Index (reverse) (reverse) 20) Sars 41mm. 20mm. _ .487 2.05 41mm. 12mm. .292 3.41 BS ova sie eae Stet Seats 6 af eA ears Apa Ses fap BES cca ws 32 15.5 .484 2.06 32 10.5 328 3.04 73 6 Se ee Aan 4h ASAE eae BAe ay ae ate eet 2 30 16.5 .550 1.81 30 BLISS .383 2.60 BOG oc 31 pL ms) . 564 sla of 4 31 14 .451 2.21 24 oe 22 2) eee 31 15.5 500 2.00 31 9 290 3.44 2 5 en See 2) |e 35 23 657 1.52 35 14 400 2.50 Sn a. 5, sce es aes Bucs = Van a 30 15.5 .516 1.93 30 8.5 . 283 3.52 BUS eae <.« 27 14 53s} 1.92 27 125 BAT 3.60 BUA ie = 35 20 aL 1S 35 12 .542 2.91 SOD... eee aoe ane sy - BUG as 3s 30 LS 083 a ar a 30 ain 366 22 3; | 7 (ae 32 IG dol 1.88 on iS 390 2.56 = | 2 eee 39 20.5 525 1.90 39 12-5 320 Se 4 B09 es... 35 eps) . 648 1.55 Sees Bh 21 ae 25 Lars, . 700 1.42 25 9 . 360 PA Uy | 21 el Ea 34 23 .676 1.45 34 12.5 . 367 2.72 51 7 26 15.5 .596 1-67 26 9 .546 2.88 Se Ss 0 = 33 Pape . 666 1.50 33 12.5 318 2.62 31 2 Se Ghee 21 36 16 444 2.25 36 4 (a Se) 319 3.13 SNES Sais « eek Nee ces 26 15 576 1.73 26 ich 423 2.36 Bac aie. ot 18 432 205 37 9 243 4.11 = sae = 38 18 473 read Ia 38 10.5 276 3.62 ie... aoe ee sbe ae 3a 26 16 615 1-62 26 12 461 2.16 32-3 37 i) 513 1.94 37 13 348 2.84 Sd 32 27 843 dies i 32 15 .468 213 2) 26 12 461 2.16 26 C5 .280 3.46 BAG. 5 os ere ae 2 ae ol 18.5 696 aed 31 14 451 BrAl SER. a's» ano. 219 .690 1.44 2S 105 .381 2.61 2 ae sara en =o) See 38 19.5 513 1.94 38 12.5 .328 3.04 RE sais « 30 20 .666 1.50 30 12.5 .416 2.40 “2 Se Heihes ee ARS See Bisse « eae S eet 233 ae 25.5 14.5 .568 1.15 25 5 8 .o13 3.18 334...°.. 3 26 as, 672 1.48 26 It .423 2.36 200 ANNALS NEW YORK ACADEMY OF SCIENCES Spirifer mucronatus mut. alpenense—Continued Adult Stage Neanic Stage Number Width Height Index Index Width Height Index Index (reverse) (reverse) 331; See 29mm. 19.5mm..672 1.48 29mm. llmm._ .379 2.63 5 Se 29 ZA 25 741 1.34 29 13.5 .462 2.14 MMOs ses 29 20 . 689 1.45 29 13 448 2.21 Be Re aT 16 -592 1.68 27 $.5 514 3.64 => eee 355 23 . 657 4252 3D 11.5 325 3.04 eo eee 32 17 531 1.88 a2 10.5 .324 3.04 o44..... 29.5 24 .813 4.22 yin} Pa 15.5 225 1:90 = ee 21 24 .888 1.122 27 14 918 12 aa hs ee 29 18 .620 1.61 29 11 ae pas 52! Hee ar : ff crate 2 ae =) | Sees 28 18 642 1.55 28 10 oT 2.80 2] Pee ee 35 16 484 2.06 30 11 ano 3.00 5-9 3S eas ry f £2) . 159 fhe! a | 14 .518 1.92 222 Se oe Soak went ee ae Seite = aie = ahaa = ee 54 19 908 1.78 s+ 9.5 .279 ope. “3 eee hats See Looe — as (ee eee 54 eae 661 41.51 34+ 14 411 2.42 ae) [on jet oe) ves = le) Or @ 04) = =| v4) Nie = oO Ho ~] i] bo iS =e =e Do 19.5 .990 1.69 DD 15 393 2.53 5 29 17 . 586 £50 29 9 310 3.22 OG es. ae 50) ee 39.5 18.5 465 2.13 39.5 11 278 3.09 14. 15.: OO 7 6) ge = a (oh) Cr OT Ye oon Ol co | be [eat] Ho ww ~~) Ww Ww feet OT oo on) jet. per He va) bo ¢ ~ | & @ iD. =f = ee 5 ea 25 a 600 1.66 25 T .280 3.97 BAN oak 5 50 12.5 416 2.40 30 8 . 266 3.75 25 ae OL 20 645 1.55 31 14 .451 2.21 MD =e Sse 30 15.5 516 1.93 30 11.5 .383 2.60 SO. Se 27 16 a92 1.68 27 12.5 . 462 2.16 5 eee 21 20.5 976 1.02 21 11 .925 1.90 = ee Sigs eee Hi Aas ee ae nica Se 3D 21 . 600 1.66 30 14.5 .414 2.41 oe ee 27 17 .629 1.58 27 11 407 2.45 B25 5.2 29 16.5 .568 1.75 29 415 .396 2.52 SG...’ - D1 18.5 .596 1.67 ol 11 .oo4 2.81 BOF tien we 25 16 . 640 1.56 25 10.5 .420 2.38 BBe: = Bo 20.5 .o81 1.70 30 12.5 .ooT 2.80- = 9) 36 19 527 1.89 36 15 .o61 2.76 ore 36 15.5 .oT4 2.66 36 10 200 3.60 = | eee 29 19 653 1.52 29 11.5 396 2.52 Number oseee -eeee “eres ee ee MOOK, VARIATION IN SPIRIFER MUCRONATUS _ _ = . Or ey aaa Stage Height Index Index (reverse) 19mm. .542 1.94 1'¢25 .486 pl 20.5 661 Bee | 15 . 656 1.53 1% 447 2123 16.5 ATI 212 15-5 . 645 1.54 19 . 703 1.42 19 .644 #.bS Ay | . 700 1.42 IG 25 .500 2.00 17.5 .486 2205 16 .615 4 762 18.5 . 525 1.89 at .538 1.85 14 .518 e792 5 .968 4°76 19.5 590 1.69 1675 . 804 1.24 di . 700 1.42 15 550 1.80 20 769 130 18.5 . Et 1.40 20 . 689 1.46 2025 . 759 oof i .625 1.60 20.5 745 ise b 20:55 745 gai Y a1 750 De So 20 . T40 2235 2625 is 1.29 25.0 . 839 1.19 23 . 884 PENS 14.5 Sot 1.86 19 .716 1.39 21 807 4.23 18.5 .616 1:62 18 . 654 2352 20. 714 1.40 18.5 .714 1.40 rg 3 .811 1.20 18 . 120 7235 23.5 S70 1.14 DAS . 750 $35 a1 651 jas Width 35mm. Wo w w& OmMmwWre DA nis bt | 21 Ww Ww Ww bo w bo CAAA to o& =I WNwwd Ww ww wy wv Ww CO 7a) bb wy bw ty bw ADDAA or) -~| -! on) on) Spirifer mucronatus mut. alpenense—Continued Neanic Stage Height Index 12. 13. 15. 9. 12 at 11 12 10 => v 11 10. 10. 14. 15 10 9 Or Ot Ol Or Cr St Ov Ov OT OU Or Or Ol Ol Cr Cr Or C1 (reverse) mm. .357 .355 .900 .413 apf 5 . 300 .458 . 444 .338 .4338 .333 .291 .4038 .414 384 .B70 .409 375 pt? . 466 .dol .480 201 - =] Qu ia?) va b> W bh ~~ Oo OO rsAaes 16 WOWwWNNNNwWwH NWN WW bo I) © SS Ww Ww ws ear wm Wwe Wd Ww vw =r) He NNWNrRNWrRFrRPRNWNNWPNNNWRPRPRPNY NR RNY WY Wb Wb 9()2 ANNALS NEW YORK ACADEMY OF SCIENCES Spirifer mucronatus mut. alpenense—Continued seal SI se Neanic Stage Number ‘Width Height Index Index Width Height Index Index (reverse) (reverse) AST we 29mm. 19.5mm. .672 1.48 29mm. 11.5mm..396 2.52 ASS. e ie 27 1OES . 709 1.38 27 9.5 .ooL 2.84 ASG Face 29 22 . 158 ese 29 aby - 586 1.70 440..... 30 19 - 633 ibe aye 30 14 .466 2.14 2 ou eae 32 20 .625 1.60 32 - 14 437 2.28 442;5..... 29 18 .620 168 29 10.5 :362 2ehG 443..... 29 1325 .465 2.14 29 10 344 2.90 444..... 22 af af: .500 2.00 22 9 .409 2.44 445..... 26.5 19 CLG ay 26.5 11.5 .433 2250) 446-468 inclusive = Spirifer mucronatus mut. thedfordense. AGO. coe 32 20 .625 1.60 32 12.5 .390 2.56 AOS s Ge: 32 Pat . 656 UR 34 oe 13.5 421 Zak cy Gi lee 28 20 .714 1.40 28 13 . 464 Fob 2 Se 28 26 928 L207 28 16 «tee eae ANS sis 'ss 35 23 657 dio 35 16 .457 2.18 AWC. Ce 32 19 .593 a Br) Bye 1s . 390 2.56 ADF. ws 32 20 625 1.60 32 13 . 406 2.46 ANDO aie ens 29 .5 20 677 1.47 20-5 12 407 2.45 ca Ge ns 25 i ry 680 1.47 25 12 .480 2.08 AG SAS 6S 29.5 24 .814 a 29.5 15S .D08 1.96 BO 28.5 22 oreahel: 1.29 2560) 14.5 .508 1.96 ASOS seco. Sif 7A | OCT 1.47 31 13 419 2.38 ASTER ees 30 PAD . 700 1.42 30 13 433 2uou Ae eos 2h 38 d Rot 486 2, 05) 38 ds 342 2.92 AGO. oe 3-9 Ws) 23 730 1-36 od es 12.5 .396 2.52 484..... 29.5 Dai ses 763 tee k 29.5 14 474 vee ht 7 oo ee 32 23 718 1.39 32 1055 Sat 3.04 486..... ree PSH cere vers 36 oa 583 Lat 36 a he 333 3.00 BO elas states syst wows sks Reet 2 iy) eae BE 20.5 661 gers & 31 . 14.5 475 pope 490. oss. 26 Ze, . 846 1.18 26 Lb55 442 2.26 2 31 215 693 1.44 = AS 20 435 2.29 21 ee 31 17.5 564 a 7 a ¢ 31 125 403 2.48 2 eee PA ape 880 g eg 25 14 560 1.78 494..... 30 Lt5 583 Led 30 13:5 450 PAR 495..... ee seer awe Bs Sete