m 1 Tfff^^^^^^B ' Committee on Publication Barton W. Evermann Chairman and Editor C. Hart Merriam David White A. D. Hopkins Lyman J. Briggs PROCEEDINGS OF THE Washington Academy of Sciences Vol. XI 1909 WASHINGTON March-December, 1909 ^' i u II 3 AFFILIATED SOCIETIES Anthropological Society of Washington. Biological Society of Washington. Botanical Society of Washington. Chemical Society of Washington. Columbia Historical Society. Entomological Society of Washington. Geological Society of Washington. Medical Society of the District of Columbia. National Geographic Society. Philosophical Society of Washington. Society of American Foresters. Washington Society of the Arch^ological Institute of America. Washington Society of Engineers. WASHINGTON ACADEMY OF SCIENCES OFFICERS FOR 1909 President Charles D. Walcott Vice-Presidents From the Anthropological Society Walter Hough Archceological Society Mitchell Carroll Biological Society Theodore S. Palmer Botanical Society J. N. Rose Chemical Society H. W. Wiley Engineers Society D. S. Carll Entomological Society A. D. Hopkins Foresters Society Gifford Pinchot Geographic Society Willis L. Moore Geological Society A. H. Brooks Historical Society Jas. Dudley Morgan Medical Society Edward A. Balloch Philosophical Society C. K. Wead Corresponding Secretary ■ Treasurer Frank Baker Bernard R. Green Recording Secretary J. S. Diller Managers Class of 1910 Class of 1911 Frederick V. Coville Barton W. Evermann Geo. M. Kober L. O. Howard Bailey Willis O. H. Tittmann Class of 1912 L. A. Bauer C. F. Marvin C. Hart Merriam Vll STANDING COMMITTEES FOR 1909 Meetings L. A. Bauer, Chairman David T. Day Leonhard Stejneger E. B. Rosa G. T. Vaughan Finance A. H. Brooks, Chairman Arthur L. Day Bernard R. Green Thomas H. Kearney George R. Putnam Rules C. K. We AD, Chairman F. W. Hodge T. S. Palmer Functions Walter Hough, Chairman Leonhard Stejneger F. W. Clarke Whitman Cross Alice Fletcher Publication B. W. Evermann, Chairman L. J. Briggs A. D. Hopkins C. Hart Merriam David White Building Geo. M. Kober, Chairman J. How.ARD Gore J. N. Rose Willis L. Moore H. W. Wiley Affiliation Bailey Willis, Chairman E. W. Parker G. K. Burgess Membership F. V. CoviLLE, Chairman J. S. DiLLER L. O. Howard Geo. M. Kober Geo. W. Littlehales Vlll CONTENTS PA or Some calcite crystals with new forms; by Waldemar T. Schaller i Linnaeus as an Evolutionist; by Edward L. Greene • .17 The Laramie Formation and the Shoshone group; by Whitman Cross ^7 Studies in the American Buprestida?; by Thos. L. Casey . . 47 The stratigraphic relations and paleontology of the ''Hell Creek Beds," "Ceratops Beds," and equivalents, and their reference to the Fort Union formation; by F. H. Knowlton. 179 The age and stratigraphic relations of the "Ceratops Beds" of Wyoming and Montana; by Timothy W. Stanton . . 239 Index 295 IX PROCEEDINGS OF THE WASHINGTON ACADEMY OF SCIENCES VoL.XI,No. I, pp. 1-16 Figs. 1-8 March 31, 1909 SOME CALCITE CRYSTALS WITH NEW FORMS. By Waldemar T. Schaller. 1. calcite from andreasberg Several specimens of calcite labeled " Andreasberg, Harz Mts.," were purchased from a local mineral dealer. One specimen shows large tabular crystals with only the base {0001} and prism |10iO|. Other specimens consist of groupings of tabular crystals which are somewhat unusual in their combinations. These tabular clacites are arranged in superimposed parallel, or nearly parallel groupings forming vertical columns several cm. long, the individual crystals being up to 13 mm. wide and to 5 mm. high. The dominant forms on these crystals are the base jOOOlj and prism {1010} but charac- teristically all of them also show the new prism j { 7.6.13.0 1 as narrow rounded faces. Very narrow faces of 5 { 0551 ] and minute faces of 2;{2131| are frequently present. Some of these crystals which seemingly consist of a single crystal show an easy parting parallel to the base and may be built up of numerous thinner tabular ones in parallel position. Associated with these calcites are numerous four- lings of harmotone. Measurements of the 9^ angle for 5{0551| varied from 78° 38' to 78° 47'; calc. = 78° 32'. The dihexagonal prism faces are bright but rounded, so that no accurate measurements could be made. They gave, however, a bright signal in the mass of reflections caused by the rounded sur- faces and the measurements given below were of this bright signal. SCHALLER The faces are seen to be in the prism zone mm' and therefore could not be scalenohedrons with complex indices. 4) meas. = 27° 34' 20 32 37 calc. {7.6.13.0} = 27° 27' Av. = 27° 31' This prism face is present on all of the crystals, of which but few, however, are suitable for measurement. The results just given were all obtained from one crystal. It seems odd that this form has not been described before — at least the writer could find no mention of it in the literature — as the specimens are supposed to come from a locality so well known for its calcites. Fig. 1 It seems hardly correct to regard these prism faces as corrosion forms or some other secondary formation. While the base and prism faces on these calcites are dull, they do not show any definite signs of etching, and though the new prism faces are rounded, they are considered as belonging to a well established form for calcite. As stated, the form occurs on all of the crystals, and all twelve of the faces are present on each crystal. They are shown in fig. 1. 2. CALCITE FROM ENGLAND The crystals here described are from a group containing individ- ual ones, the largest of which measure 4 cm. by 2 cm. The exact locality is not known. The crystals have a trigonal habit due to the predominance of a steep negative rhombohedron. The forms present on these_crystals are: r\ 1011 | , M { 4041 | , «• | 5051 \ , y { 60(31 \ , o.{ 11.0.11.2!, e{0li2|, /i0221|, ^10881}, C' {0.13.13.1 } , SOME CALCITE CRYSTALS WITH NEW FORMS 3 w{4153i, G;!7295|, 2^13145}, z;{2131 } , r j 5382 | , )' | 3251 | , and a new form p | 4.20.24.7 } . The following table gives the average of the measured angles compared with the calculated value. SYMBOL NO. MEAS. MEASURED CALCULATED LETTER 0 P •^ P M 4041 3 0°02' 75° 48' 0°00' 75° 47' «. 5051 2 0 02 78 11 0 00 78 32 O' 11.0.11.2 1 0 12 79 35 0 00 79 33 y- 6061 2 0 02 80 11 0 00 80 24 e 0112 2 0 00 26 15 0 00 26 15 f 0221 5 0 00 63 07 0 00 63 07 d 0881 4 0 00 83 01 0 00 82 46 C- 0.13.13.1 1 0 12 85 30 0 00 85 32 n 4153 1 10 55 56 22 10 54 56 26 G: 7295 3 12 22 58 22 12 13 58 14 w 3145 2 14 25 35 05 13 54 35 25 V 2131 2 19 07 69 10 19 07 69 02 r 5382 8 21 43 74 28 21 47 73 51 y 3251 2 23 21 77 02 23 25 76 64 P 4.20.24.7 13 8 56 72 09 8 57 72 20 The indices of the form above referred to as 1 4.20.24.7 | could not be definitely determined. Many faces of this form are present in these crystals, in fact, the form is a characteristic one for this group as it is present on all the crystals. But it is always rounded and uneven and will give when mounted on the goniometer, either a group of reflec- tions or else two or three distinct signals. The various measurements for this form are given below. SCHALLER 6 as H o -< S Calc. 0 = Calc. p = < IB ft. o 2 j 8° 57' 72° 20' V IC 6 P- 8° 17' 72° 08' signal distinct. 1 minute 2 • P- 8 20 71 36 signal very faint. [f. 8 23 71 40 signal distinct. [ P- 8 20 72 08 signal very faint. 1 small 3 1 P- 8 23 71 15 signal very faint. [f. 8 48 71 59 signal distinct. fP- 8 26 71 31 signal very faint. 1 small 3 ^P- IP- 9 01 72 19 signal very faint. 8 23 71 40 signal distinct. 1 ^m all 3 /P- 8 46 71 05 signal distinct. oillctil \P- 9 06 72 19 signal distinct. 2 mi"nii+fi 1 /P- 8 54 72 15 signal very faint. ^ iililiU. tc \f- 9 13 72 17 signal distinct. 2 sm3.ll Q /P- 8 52 71 58 signal very faint. o \P- 8 15 71 30 signal very faint. 9 Qm all 1 /f- 9 56 72 20 many signals. ^ olilctii \f- 10 19 73 23 signal very faint. fP- 9 24 72 04 signal very faint. 2 small 3 1 P- 10 17 73 43 signal very faint. [f. 9 16 72 41 signal distinct. 2 minute 2 P- 8 37 71 48 signal very distinct. fP- 8 44 71 56 signal distinct. 3 small 2 P- 9 23 72 38 signal distinct. [P- 9 07 72 38 signal distinct. 3 small 2 P- 8 34 71 58 signal distinct. 3 minute 1 P- 9 15 72 36 one distinct signal. 3 small 1 P- 8 48 72 51 signal distinct. Average of meas. of 13 faces 8° 56' 72° 09' An average of the values with "signal distinct" is 8° 45' 72° 11' or nearly the same. While these measurements are near to the angular values of the simple form 1 15G2, | which are ^ = 8° 57', p = 70° 00', the p angle is suf- ficiently different to preclude their reference to this form. If we take the average measured value ^ = 8° 56', p = 72° 09', we find that the ratio /i :^ =1 :5 (^ calc. = 8° 57'), while the ratio i.i =3.392 which gives as the indices 1.5.6.1f or 4.20.24.7, which are not especially com- plex and the calculated angles for which are ^ = 8° 57', p = 72° 20' SOME CALCITE CRYSTALS WITH NEW FORMS 5 agreeing fairly well with the average of the above given measure- ments. The above detailed description is given in order to call atten- tion to these trigonal calcites in the hope that some one would find a similar group having this new form which would admit of accurate measurement and thereby definitely determining the symbol of this form. Fig. 2 Fig. 3 A single minute face of C- {0.13.13.1 1 was noted and this was poorly developed, giving a poor reflection. Sansoni^ gives this form ' Zeit. Kryst. Min., x, 564. »Min. Pet. Mitth. xii, 487, 1891. (Ref. Z. Kryst., xxiv, 431.) 6 SCHALLER as new and it is also described by Hofer^ as new, being determined by zonal relations; but it is considered as doubtful by Goldschmidt,^ he suggesting that the correct symbol may be {0.14.14.1 | though San- soni's measurements, like the writers above, agree well for 1 0.13.13.1 1 . Fig. 2 shows one of the smaller crystals of this group with the impor- tant forms developed at one corner. As shown in this figure, the combinations vary at similar crystallographical places, being ofttimes much simpler at some corners than others. While all the intersection lines are here drawn straight, most of them are considerably curved on the crystals, this being particularly true of the larger crystals, which, moreover, show a much simpler combination than the smaller ones. Fig. 3 is an attempt to show the actual appearance of these larger crystals. 3. CALCITE FROM CALIFORNIA The crystals here described were collected by the writer some years ago. They occurred in cavities in what seems to be an altered rock near Fort Point, San Francisco. They were found in the same locality which has furnished pectolite and datolite^ as well as apophyl- lite and gyrolite^ though the crystals here described were not found directly associated with any mineral other than quartz. The crystals are of the common scalenohedral v habit of calcite with e and r as termination, and often reach a size of several cms. They differ some- what from the ordinary calcite crystals, in that the form T { 5382 } is almost always present as a very narrow face below v and minute faces of A/" 1 4.16.20.3 | are also almost always present. One single crystal, colorless and transparent, and about 4 mm. in diameter showed some differences from the other crystals and was therefore measured. The essential difference was that the e faces were replaced by two faces of 3':{2358i, the e faces being entirely absent. In addition a new form in the zone r e was noted. The forms present on this_crystal are: r 1 1011 } , •z;|2131|, r {5382 | , y: | 2358 } , N { 4.16.20.3 } , and the new form fe { 1 .6,7.13 \ . The average measured angles are compared with the calculated one in the following table: •Index. Kryst. Form. I, 388. * Bull. Dept. Geo!. Univ. Cal., ii, no. 10, 1901. » Bull. U. S. Geol. Survey, S6S, 124, 1905. SOME CALCITE CRYSTALS WITH NEW FORMS MEASURED CALCULATED LETTER 0 P <^ P V 2131 2 19° 04' 69° 04' 19° 07' 69° 02' r 5382 2 21 42 74 13 21 47 73 51 y: 2358 6 23 22 28 19 23 25 28 15 N 4.16.20.3 4 10 55 80 32 10 54 80 35 ft 1.6.7.13 3 7 31 26 27 7 35 26 27 Fig. 4 Only three faces of the new form are present. They are all nar- row and bright but striated like almost all the faces in this zone. This striation caused the reflections, particularly the first one, to broaden out so that accurate measurement was not possible. The three meas- urements are: SCHALLER calc. for 1 1.6.7. 13} meas. (1) (2) (3) 7° 35' 7 67 7 04 7 01 p = 26° 27' = 26 34 = 26 24 = 26 24 The form is considered as well established. It is shown in fig. 4 which illustrates the ideal combination found on the crystal measured. 4. CALCITE FROM NEW MEXICO These crystals were received from Mr. C. H. Gordon, the locality being Macey's Mine, 2| miles east of Hillsboro, New Mexico. They are in loose crystals, a centimeter thick and somewhat longer and are associated with large endlichite crystals. The calcites often have partially adhering white or pink gangue. Almost all the crystals, usually doubly terminated, show only the prism w 1 1010 | and the negative rhombohedron e {0112 | . The e faces are striated parallel to their intersection with r while the prism faces are covered with triangular markings or etch figures. These triangular markings always point toward the e face, the base of the triangle being nearest the r face. Two crystals showed considerable differences from the others and were therefore measured with the results given below. Crystal no. 1 is prismatic, 12 mm. long and 4 mm. thick, is colorless and trans- parent. It is shown in fig. 5 which represents the crystal as it actually is, excepting that the narrow scalenohedra in the zone r e are not shown. The forms present on this crystal are: m { 1010 } , a 1 1120 | , r|10]l|, M{4041|, w I 28.0.28.1 1 , e | 0112 j^ic; | 3145 } , G { 5279 } , / 1 2134 \,v\ 2131 1 and the new form 1 1 3.10.13.3 \ . The following table shows the angular values : MEASURED CALCULATED 0 P 0 P r 1010 3 0°01' 44° 36' 0°00' 44° 36' M 4041 3 0 01 75 44 0 00 75 47 CO 28.0.28.1 3 0 01 87 48 0 00 87 55 e 0112 1 0 05 26 16 0 00 26 15 w 3145 1 14 08 35 20 13 54 35 25 G 5279 2 15 56 34 27 16 06 34 23 t 2134 1 19 55 32 57 19 07 33 07 V 2131 4 19 06 69 07 19 07 69 02 I 3.10.13.3 2 12 56 75 33 12 44 75 31 SOME CALCITE CRYSTALS WITH NEW FORMS 9 For the new form I { 3.10.13.3 | only two faces arc present both of which gave poor reflections, one much more so than the other. The one giving a poor but distinct signal, gave: 0=12° 54' (calc. = 12° 44') io=75 36 (calc. = 75 31) The limits of repeated measurements of these two faces gave: calc. for {3.10.13.3} 0 = 12° 44' 12° 42' - 13° 03' 12 24 - 13 17 p = 75° 31' 75° 30' - 75° 36' 75 29 - 75 36 Average 12° 56' 75° 33' The form was seen to lie in the zone rm and measurement of the angle (1011) : (13.10.3.3) gave 50° 35' (calc. 50° 16'). The form, considered as well established is shown in fig. 5. Fig. 5 Crystal no. 2 is incomplete, and as now formed, nearly equidi- mensional, measuring a centimeter in diameter. It is colorless and while transparent in part, is cloudy and translucent in other places. The forms present on this crystal are: m\ 1010 \, r\ 1011 1,^1 0jl2 | , /j 0221 1, and the three new forms o { 0.14.14.17 | , O | 8.2.10.3 } , lO SCHALLER q I 2.16.18.1 } of which the latter is doubtful, are given in the following table: The angular values NO. MEAB. MEASURED CALCULATED 4> P ^ P r 1011 1 0° 29' 44° 56' 0°00' 44° 36' e 0112 3 0 00 26 15 0 00 26 15 f 0221 2 0 00 63 33 0 00 63 07 0 0.14.14.17 3 0 00 39 02 0 00 39 05 0 8.2.10.3 2 10 29 70 58 10 54 71 38 q 2.16.18.1 4 5 43 86 33 5 49 86 36 The large faces of 0| 8.2.10.3} are dull and give no distinct signal, only a blaze of light. Two faces were measured with the fol- lowing result: calc. (j> = 10° 64' meas. (1) = 10 21 meas. (2) = 10 36 p = 71° 38' meas. (1) = 71 07 meas. (2) = 70 48 Although the faces are dull and give no distinct reflections, the meas- urements serve to fix the indices and as the form occurs in large plane faces it is considered as well established. The corresponding negative scalenohedron .S { 2.8.10.3 | has been noted a number of times. Measurements for the form o {0.14.14.17 j are as follows, the faces being bright and giving fairly good reflections. The angles between the three faces of this form and e were measured. 0.14.14.17 : 0112 12° 52 - 12° 53' p= 39° 07' - 39° 08' 12 44 - 12 45 = 38 59 - 39 00 12 43 - 12 44 = 38 58 - 38 59 Average (^) = 39° 02' limits, 38°58' - 39° 08' The only form for which the angle agrees is the complex one {0.14.14.17}. Two simple ones which approximate to {0.14.14.17} are given in the table below, but as can be seen neither one is sufl5- ciently close to the measured values. calc. for {0556}, = 39° 25' calc.for {0.9.9.11} = 38 54 SOME CALCITE CRYSTALS WITH NEW FORMS II Four faces of the form j 2.16.18.1 | are present, the remainder of the crystal having cleaved off. While all the faces are bright, they are very rounded and in some cases form with in \ 1010] a rounded surface. They gave, however, a bright signal in the midst of the mass of signals caused by the rounded surfaces and it is these single bright signals that were measured. calc. 0 = 5° 49' p = 86° 36' meas. 5 48 86 39 5 31 86 27 5 54 86 33 5 37 86 33 Av. - 6 43 86 33 Fig. G Measurements of the prism face (in zone cem) gave p = 89° 58' instead of 90° 00'. While the above measurements agree very well with the calculated angles, the form cannot be considered as determined, as it is too rounded. It must be classed as a doubtful form. The forms are shown in fig. 6, in which it is attempted to show the crystal as it actually is. 5. CALCITE FROM RHODE ISLAND The crystals here described were received from Mr. A. L. Ste- venson of Providence, R. I., and the locality is given as Dexter Lime Quarry, Lincoln, R. I. To the kindness of Mr. Stevenson in sending the specimens the writer's thanks are due. The crystals are on lime- stone and are associated with a pearly mineral forming almost color- 12 SCHALLER less spherulites which have become brown on the surface. Other specimens show "nacrite," a talcose or micaceous mineral, quartz in flattened crystals, and curved rhombohedra of dolomite. The calcite is found in three types of crystals: (1) simple rhombohedra r{ 1011 } , abundant. (2) crystals apparently prismatic and more fully described below. (3) scalenohedral crystals showing large faces of t; | 2131 S , but the crystals were so rounded that no measurements were made of them. Those of the second type are found on the rhombohedra of the first type and are not over several millimeters long while the sim- ple rhombohedra are seldom over a millimeter thick. While the rhom- FiG 7. bohedral crystals are usually smooth and bright, those of the second type are very much etched and rounded, making it difficult to obtain satisfactory readings. The crystals have a prismatic appearance due to the presence of the negative rhombohedron ?r| 0.17.17.1 } and a form vicinal to a { 1120 | and having the approximate symbol t I 17. IG. 33.1 j . No prism faces were noted on the single crystal meas- ured. The forms present are: e 10112}, £{0111}, ?r | 0.17.17.1 1 , S: 1 11.8.19.3 ! ?, t i 17.16.33.1 } ?. The faces of most of these forms were so poor that the correct symbols could not be definitely deter- mined. Fig. 7 shows the general appearance of the crystal except that while the intersection of two faces is here shown as a straight line on the crystal they are frequently curved and rounded. The SOME CALCITE CRYSTALS WITH NEW FORMS 1 3 crystal was set in polar position by means of the e faces and the other forms gave the following measurements: fiOlll!. p calc. = 44° 36' meas. 47° 53' " 45 45 42° -45° All faces gave a mass of signals. !rJ0.17.17.l! p calc. = 86°35' meas. 87°07' 86 12 86 45 86 16 86 30 86 approx. All faces gave merely a mass of brightness and the angles measured repre- sent the center of this mass. S:|11.8.19.3|,calc. P P £ OlTl 0° 02' 44° 37' 0°00' 44° 36' t 2134 19 23 33 07 19 06 33 07 6 1232 19 06 51 34 19 06 52 32 tJ 11.5.16.21 17 39 33 43 17 47 33 40 Fig. 8 SOME CALCITE CRYSTALS WITH NEW FORMS IS The negative unit rhombohedron e | 0111 1 , a rare form, is present as three faces, all very small, giving poor reflections. The new form shows only one face, a very narrow one, not striated and giving a fair reflection. It is very close to / { 2134 } with which it occurs. The signal is slightly blurred and measurements of the two sides of the signal gave: 9& = 17° 25' to 17° 53'; /o = 33° 42' to 33° 44'. calc. = 17° 47'; calc. = 33° 40'. While the indices are somewhat complex, if we change them to the orientation given by Goldschmidt {G^ in his Winkeltabellen the indices become { 7297 | and the form fits in well between two known forms. 6282 (G2) 7297 new form. 8.2.I0.8 (G,). Though occurring but once the form is considered as well estab- lished. The crystal is shown in fig. 8, on which, however, the new form is not drawn, but its position can be easily determined as it lies very close to / { 2134 | . SUMMARY The new forms for calcite described in this paper are grouped together in the following table, the coordinate angles and p being given for each form. Well established new forms NO. BTMBOIi P 1 J 7.6.13.0 27° 27' 90° 00' 2 0 0.14.14.17 0 00 39 05 3 0 8.2.10.3 10 54 71 38 4 ft 11.5.16.21 17 47 33 40 5 k 1.6.7.13 7 35 26 27 6 I 3.10.13.3 12 44 75 31 1 6 SCHALLER Three doubtful forms are also described. 1 p 1 4.20.24.7 1, probably correct symbols, but considered doubtful because the faces were too uneven to allow of satisfactory measurement. 2 q {2.16.18.1}, a doubtful form too rounded for definite deter- mination. 3 t 117.16.33.1}, a doubtful form too rounded and uneven to allow of definite determination. Chemical Laboratory, United States Geological Survey. PROCEEDINGS OF THE WASHINGTON ACADEMY OF SCIENCES Vol. XI, pp. 17-26 March 31, 1909 LINNiEUS AS AN EVOLUTIONIST.^ By Edward L. Greene. Not more than two decades have passed since with most people who had interested themselves in such matters, and with quite all who had not, evolutionistic theory and Darwinism were synonymous; the supposition being that Charles Darwin had been the original inventor, as well as the strong promulgator, of the hypothesis of the descent of present-time species of living things from earlier types. That mis- understanding nowhere now prevails; and while a multitude of talkers and writers on all sorts of topics use freely the term evolution, Darwin- ism is less frequently mentioned; for it is coming to be realized some- what generally that there were "Darwinians" not a few, not only be- fore the Darwin of the nineteenth century, but even before that almost as remarkable grandsire Darwin of the eighteenth. There were evo- lutionists among the Greeks of five and twenty centuries ago, and even among the earliest luminaries of Christian philosophy and the- ology of a period only less remote; while after the revival of learning, and of an interest in nature study, evolutionistic ideas found expres- sion not infrequently: and of late, historians of science are bringing all this to light. The catalogue of more or less distinctly evolutionistic naturalists who lived before the end of the eighteenth century, and who gave some expression to their ideas on this topic, is not a short one; but the name of Linnaeus has not, in so far as I can learn, been placed on that list hitherto, except very hypothetically.^ ' Read before the Biological Society of Washington, November ii, 1905. ^ In the environment of the idea of evolution Linnaeus may be considered not as a positive but as one of the negative factors. — Osborne, From the Greeks to Darwin, p. 128. l8 GREENE For any possible expression of views as to the origins of groups of plants and the permanency or mutability of such groups, one would naturally look, not to his many volumes of taxonomic and descriptive writings, but to just such a work as the Philosophia Botanica. Yet there one looks in vain for any expression that is not positively and unmistakably contrary to the idea of evolution. In respect to the origin of genera, that which he says — and with Aristotelian brevity and conciseness — is this: "Every genus is na- tural and was in the begininng of things created such."^ And be- cause of this — which might well enough be called the supernatural rather than the natural origin of genera — because of this origin, he argues that: "No one genus is ruthlessly to be divided and treated as if there were two; neither are any two or more to be put together as if constituting only one." In the light of such a pronouncement, one could not attribute to Linnaius any notion of the gradual evolution of such groups of species as constitute genera; and if a genus is to have such origin, so, by the necessities of logic, are species also made; and he says: "All species are certain diversities of form which the Infinite Being created so in the beginning; which forms according to immutable laws of generation, produce always their like." From this he proceeds to establish more firmly, if possible, the immutabihty of species by defining generation as being the actual "continuation of the species;" and he concludes by calling attention to how, as by necessity, this origin of all species precludes the possibility of any new species ever arising. And thus, under the heading of species does our author seem to have builded even a more insurmountable wall against the possibility of one's successfully claiming him for the camp of the evolutionists. There remains one other category of plant forms, of lower rank than species, recognized by Linnaeus; that of varieties. Unless I err, he claimed that he had been the first of systematists to recognize varieties and to teach the distinctions between variety and species. Will he so define variety as to leave an opening for the possible development of a species out of that which started forth at first as a mere variety? If we use our own reason, and credit Linnncus with not momentarily forgetting to use his, we may not look to see him contradict himself * Genus omne est naturale, in primordio tale creatum. Linn. Syst. Nat. and Philos. Bot. LINNiEUS AS AN EVOLUTIONIST I9 quite so promptly. He has said, and that in the paragraph next pre- ceding the definition of variety, that all species — not most of them but all of them — were constituted such by the Creator in the very begin- ning of the existence of plant life and form. He will not subvert this proposition; at least, not in the very next sentence. His notion of a variety is, that it is such alteration of a species as may have been in- duced by changed conditions of climates, soil, temperature, exposure to or shelter from high winds, or any such items of mere environment; and he does not fail to add that, on the restoration of the plant to its old environment, it reverts to the original type form. One sees at a glance that this is not our twentieth-century idea of a botanical vari- ety; but it is the Linnaean idea, and with that alone we are here con- cerned. The man makes so small account of varieties, from the taxo- nomic point of view, that he concludes his discussion of the topic with an apology for giving them place and mention in his books of systematic botany. "Variation," he says, "is in such matters as the size of the plant, doubleness of flower, a crisped or curled foliage, a difference of color, odor, flavor, etc." But he adds: "Many varieties of plants are in favor with gardeners, and agriculturists, others with florists, while still others are in esteem with pharma- cists. " From these expressions it is plain that Linnaeus did not con- sider these changeable and even transient forms worthy of any serious consideration by botanists proper, and admitted them to his books only as in condescension to the wants of those classes of tradespeople whom he mentions. It may here be added that in almost all more recent botany, varieties, such as Linnaeus had in mind when he wrote the defmition, find no place. One looks for the account of them, if anywhere, in the calendars and catalogues of gardeners, pomologists, nurserymen and florists. I have long understood how very definitely and absolutely this fine book, the Philosophia Botanica, excludes every idea of a possibly evolutionary origin for any species of plant. And yet, Linnaeus was an evolutionist. Nor is this so passing strange, in a world where men in great numbers — even some of high standing and great ability — say one thing, and think the very opposite. That he entertained doubts as to the truthfulness of the proposition that everything that ought to be called a species had been made as it is in the beginning, is a discovery that I made quite fortuitously. 20 GREENE In the study of some species of Thalictrum I had need to consult a certain page of the Species Plantarum. Reading his account of T. fiavum, and next below it that of T. lucidum, his concluding note regarding the species last named quite startled me. His Latin sen- tence here, as in many another place, is highly figurative, quite after the style of many a classic rhetorician and poet; and I read it again, and very carefully, to see if the idea which the first reading conveyed to my mind was quite that which the author meant to convey. There could be no doubt. Putting it into plain English prose; making it read as one would now write the same thought, his note on Thalictrum lucidum is this: "The plant is possibly not so very distinct from T. flavum. It seems to me to be the product of its environment. "^ As helping toward a full understanding of this pregnant remark it must be said that the sy>q.c\q.s flavum inhabits the cool moist meadows of northern Europe, while lucidum belongs to southern France and to Spain. Each has then decidedly its own environment. Each was known to be equally established as a permanent and indigenous plant form. Linnaeus's reason for naxamg flavum as the parent and luci- dum as the offspring, was a reason no better than this. T. flavum was of his own northern country and he knew it well. T. lucidum was a southerner, and he was less familiar with it; probably had never seen it but in a northern garden. That was all. It was a thing far enough from being amenable to his definition of a variety. It seemed a species; yet he doubted that it was any more than a daughter species to Thalictrum flavum. The one had been created a species in the beginning, the other was probably not so old; more likely to have come into existence away down among the more arid hills of Spain; but it had come to stay. Rather many plant forms that had been reckoned good species before Linnaeus and that are now again so considered everywhere today, were with Linnasus mere varieties of other species. But he declined so to treat Thalictrum lucidum. If the relation between this denizen of the fervid South and his plant of the frigid Scandinavian peninsula should be declared nothing more than the relation between a specific type and its variation, botanists would be asking how long before he would make an end of species altogether. He was not himself convinced that it was a mere variety, * Planta, an satis distincta a T. flavo? videtur temporis filia. Species Plantarum, i Ed., p. 547; 2 ed., p. 770. LINN/EUS AS AN EVOLUTIONIST 21 and so he retains it as a probable species, yet to his half secret thinking not at first created such, but the descendant of another species. Famihar as I had been for many years with the Species Plantarum as a book of reference, this one discovery upon which I had now stumbled, seemed so much like a new revelation of the mind of Lin- naeus that within a very few days I had read every one of the 1682 pages of the edition of the year 1764 in search of other kindred expres- sions regarding the possibility of the descent of some species from others. Only three pages away from the record of his thought about the origin of the Thalictrum, under Clematis niaritiina occurs this remark: "Magnol, and also Ray have adjudged this to be a variety of C. Flammula. I should rather think it is derived from C. recta under altered conditions." Now while this remark, standing by itself, might indicate an opinion that the plant under discussion was a mere variety of Clematis recta, yet Linnaeus did not so place it in this or any other of his books. He gives it the rank of a species, distinctly, and must needs have done so in view of his own definition of varieties as transient forms, developed mostly under cultivation. Clematis maritima, as its name indicates, is a seaside species, unchanged in its character from immemorial ages. He knew all this and held it to be not a variety but a derivative species; not one so created in the begin- ning. Again, next to the familiar Achillcea Ptarmica, of almost all Europe, he places the name and description of Achillcea alpina known only from the mountains of Siberia. No botanical authority has ever seemed to think of this as possibly a mere variety of A. Ptarmica of Europe; no more does Linnaeus; but while according it full specific rank, and as if forgetful of all he had said in the Philosophia Botan- ica upon such matters, he appends to his technical account of A. alpina this most evolutionistic suggestion: May not the Siberian mountain soil and climate have molded this out of A. Ptarmica?^ Among the more elegant flowering plants adorning the borders of subsaline marshes southward in the United States is one which Lin- naeus denominated Hibiscus Virginicus!^ It is exclusively North * An locus potuerat ex praecedenti formasse banc? Species Plantarum, 2 Ed., p. 1266. * Kosteletzkya Virginica of more recent authors. 22 GREENE American, and even here of somewhat restricted range. A similar species, of distribution as limited and peculiar, belongs to southern Europe, inhabiting the shores of the Adriatic Sea. Now between these two kinds of Kosteletzkya occupying widely sundered conti- nents, and neither one much more than local, each along its own little line of seaboard — between these two Linnaeus apprehends the exist- ence of a more intimate relationship than the most advanced evolu- tionists of the twentieth century would be likely to affirm. He re- marks a very close superficial likeness between them; so close that, were that all, he would declare them to be specifically one and the same; but, in the characters of their little seed pods or capsules they are so unlike that on this account separate specific rank must be ac- corded both, and so he places them; concluding, however, with this thoroughly evolutionistic query: May not the Venetian species have sprung from the Virginian?^ The more probable theory of the evolutionist of our time would be, that both are descendants from some common ancestor that had a more general distribution and is now extinct. But, that Linnaeus was disposed to regard the Virgin- ian species as having been created such as it is, and the Venetian as having originated from that in after times, is enough to warrant our regarding him as an evolutionist. I shall cite but one more instance of Linnaeus's tacit acceptance of species as derived from other species through altered environment. The case is that of the cultivated beet. The genus Beta, in his view, consists of two species only, one the Beta maritima indigenous to Old World seashores, a wild plant altogether, and never under cultivation, and, in this wild condition not given to variation, but always one and the same thing. The second species is Beta vulgaris, one not known as a wild plant anywhere, but existing from immemorial ages in gar- dens and fields as a cultivated plant, and that under many marked varieties. Now the short and easy method of dealing with a genus like this — a method many an indifferent systematist would follow — would be to make the guess that, as only one wild species is known, all the cultivated things of that genus are but so many varieties of the one species. The whole tendency of Linnaius's mind was in this direction, that is, of reducing both genera and species to a minimum. ' Species Plantarum, 2 Ed., p. 981. LINNJEUS AS AN EVOLUTIONIST 23 But there was a difficulty here with these two members of the genus Beta, the simple and unvarying wild kind, and the extremely variable one of cultivation. The cultivated plant was hardy, often ran wild, as it were, by escape from cultivation; but these reverts never were found to be equivalent to Beta marilima or anywhere near it. The Beta vulgaris self-sown and run wild for years, and greatly altered from its cultivated condition, yet invariably retained a character of its own; so that no one would think of calling it Beta tnaritinm; therefore, with Linnaeus the collection of the varieties of cultivation must be admitted as forming a distinct species of which the native original was unknown, and probably long ages ago extinct. To this view of the case he was perhaps inclined; yet not so strongly as to preclude his offering, in a note, this very different suggestion: "Possibly born of Beta maritima in some foreign country.'"* The force of this alter- native proposition will be lost to any one who does not recall that, according to the Linnaean account of a variety. Beta vulgaris if it originated from seed of Beta maritima originated not as a variety but as a species; and such an origin as he thinks the cultivated beet may have had from the wild one would amount to nothing less than what is now called a mutation : one of those sudden leaps or transitions from one thing to another which we have been learning to take into account only lately. A like instance confronted Linnaeus under the genus Cynara, the type of which genus is the true artichoke, and has been culitvated from no one knows how far anterior to all written records. Under this old type species, Cynara Scolymus, Linnaeus admits three marked varieties. Then he proceeds to name and define a second species, a very distinct one, but with a well authenticated history as having arisen nd come into existence as a seedling of the other species. He intimates that he would have liked to be able to consider it a hybrid,® but as its parentage as a hybrid could apparently lie nowhere but between two of the three varieties of the other species, the fact would remain that it was a species derived not from two parent species but from one alone. It was another of those abruptly derivative species in which Linnaeus was disposed to believe despite those hard half- theologic definitions of his Philosophia Botanica. * Species Plantaruni, 2 Ed., p. 322. ° Species Plantarum, 2 Ed., p. 1159. 24 GREENE In the progress of these enquiries into the mind of Linnaeus as to the origins of species nothing that I have come upon has more deeply interested me than his remark upon the two species of sundew com- mon in northern Europe, Drosera roiimdifolia and D. longifolia. They are very pecuHar plants, uncommonly interesting from several points of view, and have in recent years profoundly engaged organ- ographers and physiologists; but Linnaeus was most interested in their ecology as bearing upon the problem of their genealogy. Both are bog plants, though far enough from being found in every northern bog. They seem to be particular about the kind of soil, the amount of moisture, the nature of the exposure, and also the plant associates amid which they will establish their habitation; and both species are at perfect agreement as to all special details of bog environment which they demand; for where one is found, there too is the other. They are much alike in size, mode of growth, degree of hairiness, form and color of flowers, etc., but the leaf blades in one are round, while in the other they are so much elongated as to be called narrowly oblong; and this one strong distinguishing mark is constant. There are no plants among them to show leaves intermediate between orbic- ular and oblong. They ought to be, and I think that by all botanists except Linnaeus, both before his day and ever since, they have been held distinct; and even he did not positively affirm the contrary, but only expressed a doubt; and the sole reason he has for doubting the validity of D. longifolia is, that it and its mate species always occur under precisely the same conditions and together.^" It is such a reason as none but a confirmed evolutionist could give; the expression, perhaps unguarded, of a mind no longer very patient of the opinion that two species of the same genus can have the same native environ- ment. A creative fiat could, of course, as readily make two species of a genus suited to certain conditions as one, and as easily twenty as two; and so no believer in the special creation of all species could have felt this doubt about the sundews to which Linnaeus gave expres- sion. It has been thought that the mind of Linnaeus as to the absolute fixity of species underwent a change between the years 1751 and 1762, though only in so far as to induce him to admit the origin of more '"Habitat ubique cum praecedente; an itaque satis diversa species? Species Plantarum, i Ed., p. 282; 2 Ed., p. 403. LINN.I^US AS AN EVOLUTIONIST 25 recent species by hybridization." My own impression is that few if any of the plants thought by Linnocus to be hybrids are at all of that origin, according to the views of modern botanists, but rather, for the most part at least perfectly distinct and genuine species. But what I have herein, I think, clearly shown is, not only that Linnaeus accepted and admitted to his books, as species, forms he thought of as devel- oped from other species, not by any crossing, but through mere environ- ment— natural environment in some instances, artificial in others. And this bent of his mind was so strong that he could scarcely admit two m mbers of a genus to be specifically distinct if found to occur always under the same physical conditions. Again: while it is gen- erous to allow to the great nature student the eleven years between 1751 and 1762 in which to have changed his views a little as to the fixity of all species, the simple fact is that nowhere were the views set forth in the Philosophia Botanica of 175 1 more squarely contradicted than in the Species Plantarum of 1753. There were two years inter- vening between the dates on the respective titles; but most likely he was engaged in writing the works, at least in part, simultaneously. But the great man was writing and publishing as other men of genius had done before him, under environment. In a letter written by Linnaeus as early as 1747, addressed to his most intimate and trusted friend, J. G. Gmelin, author of the Flora Sibirica, he gives confidential expression to the restraints under which he feels that he is obliged to write on matters that impinge upon the domain of theology; to his unwillingness to face the disapproval of the Lutheran and orthodox ecclesiastics who, in his day, ruled the destinies of all seats of learning in Sweden. He says to Gmelin: You disapprove my having located Man among the Anthropomorphi- But man knows himself. Now we may, perhaps, give up those words. It matters little to me what name we use; but I demand of you, and of the whole world, that you show me a generic character — one that is accord- ing to generally accepted principles of classification — by which to distin- guish between Man and Ape. I myself most assuredly know of none. I wish somebody would indicate one to me. But, if I had called man an ape, or vice versa, I should have fallen under the ban of all the ecclesiastics. It may be that as a naturalist I ought to have done so.^^ " Osbom, From the Greeks to Darwin, p. 129. "This, though written as we have said in 1747, was never pubhshed until 1861. The original Latin text of the letter occurs in " JoannisGeorgii Gme- 26 GREENE The good orthodox Lutheran ecclesiastics that ruled the Swedish university in every department of it would be thoroughly content with the pronouncements of the Philosophia Botanica; and that was a book any scholar would read with pleasure and with profit; but noth- ing like that could be said of the Species Plantarum. Here, at least, in footnotes, or even in places more obscure, very briefly, veiled in figures of rhetoric, and even under the further protection of question marks, he could express his profounder convictions and feel secure. And he was secure, indeed. lini, Reliquiae quae, supersunt commercii epistolici cum Carolo Linnaco Alberto Hallero Guilielmo Stellero et al., Floram Gmelini Sibericam ejusque Iter sibericum potissimuni concernentis, ex mandato et sumtibus Academiae scien- tiarum Caesareaj Petropolitanae publicandas curavit Dr. Guil. Henr. Theodor Plieninger; Stuttgart, 1861," p. 55, and is as follows: "Non placet quod Hominem iter anthropomorpha collocaverim; sed homo noscit se ipsum. Removeamus vocabula, mihi perinde erit, quo nomine utamur; sed quaero a Te et Toto orbe differentiam genericam inter hominem et Simiam, quae ex principiis Historiac naturalis. Ego certissime nullam novi; utinani aliquis mihi unicam diceret. Si vocassem hominem simiam vel vice versa omnes in me conjecissem theologos. Debuissem forte ex lege artis." PROCEEDINGS OF THE WASHINGTON ACADEMY OF SCIENCES Vol. XI, No. i, pp. 27-45. March 31, 1909. THE LARAMIE FORMATION AND THE SHOSHONE GROUP. By Whitman Cross. * A recent discussion as to the origin and definition of the term Laramie, by A. C. Veatch,^ calls attention once more to a vexed question in American stratigraphic nomenclature. The original observations, by Mr. Veatch and others, which have led to this dis- cussion of terminology are certainly of great significance and must receive full recognition, but they do not justify, in my opinion, the proposition made by Mr. Veatch for the future use of the term Lara- mie. Another course will be advocated in the following pages. Mr. Veatch announces the discovery of an "unconformity which in the vicinity of Carbon (Wyoming) and to the southeast separates all the Laramie beds studied by the Hayden and King parties from the underlying Cretaceous." The unconformity may, he believes, involve the "whole sedimentary series of the region, or over 20,000 feet of strata." This discovery and others recently made in Wyom- ing and Colorado^ demonstrates that the similar unconformities hitherto known in Colorado and Montana are not of such local ' Published with permission of the Director, U. S. Geological Survey- ^ On the Origin and Definition of the Geologic term Laramie. Amer- Jour. Sci., 4th ser., Vol. XXIV, 1907, pp. 18-22. Jour. Geol., Vol. XV, 1907. pp. 526-549. ^ Veatch, A. C. : Coal and Oil in Southern Uinta County (Wyo.), U. S. Geol. Survey, Bull. 285, Contributions to Economic Geology, 1905, pp. 332, 335. Veatch, A. C: Coal Fields of East-Central Carbon County (Wye), U. S. 28 CROSS significance as some geologists and paleontologists have been inclined to assume. The new observations are of special interest as affecting beds of the Laramie Plains, from which a considerable portion of the supposed "true Laramie" flora has been obtained. The point emphasized by Mr. Veatch in the cited articles is one of nomencla- ture— as to the importance of the newly discovered stratigraphic relations of the Carbon section in determining what shall be the future use of the term Laramie. He attaches first importance to those relations, while I wish to express my conviction that they are, under the circumstances of this case, of subordinate importance. Mr. Veatch believes it necessary to tie the definition of a geologic formation to a type section and argues that the geographic name applied to the formation should refer to that type section. While it is no doubt desirable to follow these rules wherever practicable, it is unfortunately a fact, of common experience to geologists, particu- larly to those working in the western United States, that in many cases the name most closely connected with the locality of a key section is preoccupied as a formation name or is unsuited to such use; and in some instances there is no name attached to the type locality. The real essentials for an adequate definition of a forma- tion appear to me to be: clearly expressed stratigraphic relations, a statement of lithologic character and paleontologic data, and refer- ence to type sections, the more the better. It is fortunate if the name can refer to a completely typical locality. If, however, the definition of a formation is clear as to stratigraphic relations and the unit with those relations is a natural one, the name becomes in time more intimately connected with the definition than with the strata of any locality, for the first known sections are not always the best. A term thus becomes denotive rather than connotive. The require- ment that the name should be derived from "the type locality" becomes unreasonable and impracticable when the unit named is a Geol. Survey, Bull. 316, Contributions to Economic Geology, 1906, Part II, Coal, Lignite and Peat (M. R. Campbell in charge), pp. 246, 250. Schultz, A. R.: Coal fields in a portion of central Uinta County (Wyo.), Ibid., p. 214. Gale, H. S. : Coal Fields of the Danforth Hills and Grand Hogback, in Northwestern Colorado, Ibid., p. 267. Veatch, A. C. : GeographyandGeology of a Portion of Southwestern Wyom- ing, etc., U. S. Geol. Survey, Prof. Paper No. 56, 1907. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 29 large formation or a group not completely represented at any one locality. Such was and is the case with the "Laramie Group." As to the origin of the term Laramie Mr. Veatch points out that Clarence King, who proposed the name and defined its application, was very careful in the choice of appropriate geographic names for formations. That is undoubtedly true in a general way, still, King was not working under any rule such as that now prevailing in the U. S. Geological Survey. Much space is given by Mr. Veatch to establishing by citations the exact application of the name Laramie Plains in the seventies, and to showing that the sections of " Laramie" beds examined by the Hayden and King survey geologists were probably all above the unconformity seen at Carbon. His conclu- sion is that *' the type locality of the Laramie is Carbon, on the Lara- mie Plains."* If the term Laramie had been in fact proposed especially for strata of the Laramie Plains, or even for a formation known by King and Hayden only in the zone traversed by the Union Pacific Railroad or the Fortieth Parallel in Wyoming, it is probable that many stratigraphers would agree with Mr. Veatch in his sweep- ing conclusion that: "strictly considered, the term Laramie, there- fore, can appropriately be applied only to the beds above the great unconformity," whatever they might think of the new restriction "and below the Fort Union. "^ But the name was not so proposed. It was introduced into literature and defined by King as a compromise term for beds known to be widely distributed from Montana to New Mexico. The statement made by King is very clear as to the desire of Hayden and himself to have a name which each could use for a great series of supposedly conformable beds, without prejudice as to age. Laramie was practically proposed as a synonym for "Lig- nitic, " but not as an exact one, for that term had been very broadly used by Hayden and others. Mr. Veatch thinks that "if merely a general term without a type locality had been desired, the term Lignitic, exactly defined for the area involved, would have served all purposes.® That is possibly true, if King and Hayden had thought so, and been willing to use it, but they wisely adopted another course. There is not the slightest doubt as to the fundamental stratigraphic * Amer. Jour. Sci., 4th ser., Vol. XXIV, 1907, p. 20. » Ibid. « Ibid. 30 CROSS relations which King and Hayden thought characteristic of the Lara- mie. Mr. Veatch has shown this by ample quotations from the definition of the former. In all statements quoted and in others which might be cited the relations most strongly emphasized in regard to the Laramie is its conformity with the underlying Creta- ceous beds. King believed the Laramie to be Cretaceous; Hayden thought it transitional between Cretaceous and Tertiary. King believed that Mesozoic sedimentation ended by reason of a great Rocky Mountain revolution and that the Laramie was separated from the lowest Eocene beds, erroneously supposed by him to be the Wasatch, by a marked unconformity. Hayden thought that Laramie and Wasatch were essentially conformable. The essence of King's definition is contained in the sentence: "accordingly .... it was amicably agreed between us [King and Hayden] that this series should receive the group name of Laramie, and that it should be held to include that series of beds which conformably overlies the Fox Hill."^ " There is no difference between us as to the conformity of the Laramie group with the under- lying Fox Hill."^ There is no reference to the Carbon or any other section of Laramie beds as typical. The name would appear to have been chosen because it was deemed the best one derivable from the zone covered both by the King and Hayden surveys, and not because of a type section. It was undoubtedly believed that the Carbon section illustrated, but in common with many others, the funda- mental relation of conformity with the underlying Cretaceous. Hayden's last statement of his position was in the introduction to Lesquereux's Tertiary Flora, as follows: "The facts as we understand them at the present time would seem to warrant this general division, viz: a marine series, Creta- ceous; gradually passing up into a brackish water series, Laramie; gradually passing up into a purely fresh water series, Wasatch."" He regarded it probable that the Wasatch and Fort Union beds were identical, in part at least. The Laramie of King's conception was a natural stratigraphic unit with a well defined base to be found in many places but with ' U. S. Geol. Expl. of 40th Parallel, Vol. I, Systematic Geology, 1S78, p. 331. "Ibid., p. 348. » U. S. Geol. and Geog. Survey of the Ter., Men. VII, 1878, p. v. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 3 1 an upper theoretical limit which might nowhere be represented, owing to the assumed pre-Tertiary erosion. It was to embrace the upper group of conformable Cretaceous sediments, deposited in brackish or fresh waters during gradual continental uplift. A long period of non-deposition and great erosion in the Rocky Mountain region was assumed by King and by other stratigraphers and paleon- tologists of thirty years ago to mark the interval between Cretaceous and Eocene sedimentation. Hayden's view as to the age of the Lara- mie was largely due to Lesquereux's opinion that the flora was Tertiary, but, as the quotation above given shows, he thought of the Laramie in its fundamental stratigraphic relation as the brackish water deposit succeeding the marine Cretaceous through gradual uplift. Although the Laramie was simple in its essential definition and conception, the strata referred to it included local deposits as well as those of wide distribution, and knowledge concerning some of these beds was very meager and untrustworthy when the group was established. It is a most natural result of detailed studies during the last thirty years that several formations at the top of the group assumed to have the relations embodied in King's definition, have been found to possess other relations. But there is still a large for- mation or group answering to the fundamental part of King's defi- nition, and to such beds it seems to me both most natural and most expedient to apply the term Laramie in future. In the Laramie Plains there are, according to Veatch, 6,500 feet of conformable Cretaceous beds above the Montana marine strata and below the break at the base of the Carbon beds. The geographic term is thus still appropriate, even if the Carbon section be excluded from the Laramie. The term has now been so widely applied and for such a long time that it appears to me unwise to drop it even if there should prove to be no true Laramie beds on the Laramie Plains, as Veatch uses that designation. Among the districts named by King as exhibiting Laramie beds in representative manner is the eastern foot hill zone of the Front Range in Colorado, where they were known also to geologists of the Hayden Survey. It was in this region that Eldridge and myself discovered that the section referred to the Laramie by both Hayden and King consisted of two parts: a lower one, conformable with the 32 CROSS marine Cretaceous, and thus corresponding to the principal element of King's definition, and an upper portion exhibiting relations at variance with that definition. With the recognition of the Arapahoe and Denver formations, separated in time from the Laramie of the conformable series by an epoch of uplift and great erosion, came the necessity of revising the application of the term Laramie. The Arapahoe and Denver beds were at first considered to be the earliest deposits of the Eocene series.*" Taking them from the Laramie made no change, I think, in the definition of the latter as the upper- most member of the conformable Cretaceous series in the Rocky Mountain region, although Mr. Veatch considers it "redefinition." Soon after the recognition of the Arapahoe and Denver forma- tions other deposits of nearly synchronous origin were discovered in discordant relations with the Cretaceous section, a notable instance being the Livingston beds of Montana. It was also found by Marsh and Hatcher that the horned dinosaurs and associated vertebrates, the so-called "Ceratops fauna," occurred in Converse County, Wyoming, in strata which appeared to occupy the position of the Laramie, in conformity with the marine Cretaceous. The presence of the same fauna in the Arapahoe and Denver beds led many geolo- gists and paleontologists to the belief that the latter must also be of Cretaceous age, and that the breaks below the Arapahoe and Liv- ingston beds were of local importance. In 1892 I reviewed the accumulated evidence concerning the importance of the Arapahoe and Denver beds and deposits correlated with them by. virtue of unconformity or fossil evidence and in con- clusion advocated: — "The restriction of the term Laramie, in accordance with its original definition, to the series of conformable beds succeeding the marine Montana Cretaceous, and the grouping of the post-Laramie lake beds described, with their demonstrated equivalents, in another series to which a comprehensive name shall eventually be given."'' This course had previously been proposed by R. C. Hills.'^ In '"Cross, Whitman: The Denver Tertiary Formation. Amer. Jour. Sci., (3), Vol. XXXVII, 1889, pp. 261-282. " "Post-Laramie Deposits of Colorado." Amer. Jour. Sci., (3), Vol. XLIV, 1892, p. 41. " Orographic and Structural Features of Rocky Mountain Geology. Proc. Colo. Sci. Soc, Vol. Ill, 1890, p. 397. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 7^;^ the monograph on the Denver Basin'' the terms Laramie and Post- Laramie were again used with this significance. If it be conceded that the interests of stratigraphic nomenclature will be best subserved by retaining the term Laramie for the beds of the conformable Cretaceous section above the Montana group it is plain that the nomenclature of the younger formations, usually separated from the Laramie by stratigraphic break or unconformity, which were included through misapprehension by King and others in the original Laramie, should also be considered in this connection. It is still my belief that the use of a new group term for these forma- tions will greatly assist in clarifying the situation. The tentative term Post-Laramie has but imperfectly served the purpose intended, and is clearly unsuitable for permanent use. It appears to be left for me to finally propose this new term. The uplift that terminated deposition of the conformable Creta- ceous section was, as it seems to me, in the first degree epeirogenic, aflfecting the Rocky Mountain region from Montana to southern Colorado, with eyery probability that its influence was equally marked far beyond the limits named, especially to the north and to the south. Certain facts of observation, such as the great angular unconformities below the Livingston and Carbon beds and the stratigraphic break of equal magnitude below the Arapahoe forma- tion, may be interpreted to indicate that pronounced orogenic uplifts of as yet undetermined amount and extent were imposed upon the greater movement. Such comparatively restricted uplifts have occurred at various times and in different districts throughout the Tertiary period. That the movement was not one of continuous uplift may be almost taken for granted. Subsidence doubtless occurred but there is no evidence of true marine conditions ensuing after the Laramie epoch, as there was after the Judith River. If subsidence produced brackish water conditions in post-Laramie time it must have been of but local extent. Succeeding or in part synchronous with the Rocky Mountain uplift there was erosion and resultant deposition in many basins or on the slopes of the rising land. For such of these more or less '^ Geology of the Denver Basin, Colorado, by S. F. Emmons, Whitman Cross and G. H. Eldridge. Mon. XXVII, U. S. Geol. Survey, 1896. 34 CROSS distinctly local deposits as may now be preserved it is proposed to provide a group name. As the beds in question are genetically connected with the uplift which gave rise to the Rocky Mountains and have a wide range through the mountain province it would be most appropriate if the group name could refer to these relations. But the use of Rocky Mountain as the name of a Carboniferous quartzite, by Dowling," precludes the application of this term in another sense. The State names Montana, Wyoming and Colorado, and all the other geographic names of more than local significance which might be considered appropriate for this purpose appear to be already in use as strati- graphic terms. Believing that it would be a misfortune to adopt a purely local term for this great group of deposits, I have followed the personal suggestion of Dr. F. H. Knowlton and selected an Indian name connected with the larger part of the Rocky Mountain prov- ince. The map published by the Bureau of Ethnology illustrating the distribution of the great linguistic families of North American Indians shows the district once occupied by the Shoshonean family to embrace the greater part of the Rocky Mountain area, with a part of the great Basin. The mountains were especially the land of the Utes and Shoshones, two tribes of this family. With this broad derivation it is proposed to apply the term Shoshone Group to the deposits which unconformably succeed the Laramie and to their equivalents, and which are overlain by the Fort Union or Wasatch beds, when they are present. It is to be clearly understood that the name Shoshone refers to no typical locality or district but rather to a province. It has no con- nection with the Pliocene Shoshone Lake of King, the deposits of which he called the Humboldt Group. In manner of derivation the term is directly comparable with "Algonkian," which comes from Algonquian, the name of another linguistic family of Indians, and of the province they inhabited. Laramie and Shoshone beds may be easily separated, under the broad general definitions thus far used, where angular imconform- ity exists between them or where a corresponding stratigraphic break can be identified by lithologic or palcontologic evidence. One '^ Dowling, D. B. : Report on the Cascade Coal Basin, Alberta, Canada Geological Survey, 1907, p. 9. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 35 or the other of these conditions exists in many locahties that have been closely studied and may possibly be realized throughout the area in which the Shoshone beds now occur. But there is at least a theoretical possibility that in some places now represented by exposures deposition was continuous from the marine Cretaceous to the Fort Union epoch. In such places Laramie and Shoshone strata come together in conformity or a transitional formation must be created. It should be kept in mind while considering the subject that this situation is not peculiar to the boundary between Laramie and Shoshone beds but may arise in any case where it is desired to separate strata preceding from those formed during or immediately after a great uplift. The line to be drawn between Laramie and Shoshone in a series of conformable and consecutive deposits must be determined at some future time, when such a series is definitely known. The criterion for the distinction may be a lithologic or paleontologic change, or as is more probable, some arbitrarily selected feature. At the present time, it seems to me that the difficulty of distinguishing between Laramie and Shoshone beds is either a general theoretical one, or a local and practical one due to a lack of knowledge concern- ing the true relations and paleontologic contents of the strata in some special district. The characterization of the difficulty as perhaps largely theoretical rests on the following consideration: The uplift terminating the widespread Laramie sedimentation doubtless caused at first a rela- tively small landmass, or perhaps several of them. Denudation of such land areas must have begun at once; erosion and deposition, either continental or submarine, were, indeed, accompaniments of uplift. But as larger and larger tracts gained some notable elevation erosion must have quickly removed a great part of the most recent deposits, if not all of them. The whole of the Rocky Mountain province may thus have been denuded of the beds which would con- stitute a transition series if preserved. In this case a complete sec- tion must be looked for in outlying areas which were but little ele- vated during the Shoshone epoch. On the Great Plains side it is clear that such beds can be found only in quite restricted districts, owing to Tertiary erosions or a mantle of later deposits. On the west, too, the areas within which transition beds can exist are few. 36 CROSS The idea of transitional deposits is, of course, not new. Numer- ous sections have been thought to represent consecutive sedimenta- tion but the investigations of the last twenty years have shown breaks in many such sections, while in no case has a transitional series been established. The Hayden survey geologists believed that Laramie and Wasatch beds were, in some places at least, connected by an unbroken series of strata. Aside from apparent conformity the existence of certain species of fresh water shells in both Laramie and Wasatch has been taken as evidence of continuity of deposition. Dr. C. A. White, whose observations during the Hayden survey were of much weight in influencing the opinion of his colleagues, again expressed this idea while reviewing the relations of the Laramie to the Denver and Arapahoe formations. ^^ The presence of identical fresh water species in the Laramie and Wasatch of any given section does not, however, prove the continuity of conditions at the site of that sec- tion during the whole time involved. In the light of observations at Golden, Carbon, and many other localities, apparent conformity between the Laramie and beds above it cannot be regarded as proof of continuous sedimentation. The only evidence which can show conformity to mean continuity of deposition in this case would appear to be gradual change in fauna or flora of the epoch in question. This evidence has not been fur- nished at any locality. Referring only to the established facts of the Denver district, a section of some other region, supposed to represent continuous sedi- mentation between the Laramie and Fort Union epochs, must contain beds representing: — 1. The long interval of uplift and erosion preceding the Arapahoe deposition. 2. The Arapahoe. 3. The interval of erosion and volcanic eruption preceding the Denver. 4. The Denver. 5. Interval between the Denver and Fort Union, No section even approximating such a record has been established. '* The North American Mesozoic. Vice-Presidential address, Proc. A.A.A.S., Vol. XXXVIII, 1889, p. 14. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 37 The Laramie may be defined as a formation of brackish or fresh- water sandstones and shales, usually quartzose and coal-bearing, which constitute the uppermost division of the conformable, consecu- tive, and widespread series of the upper Cretaceous in the Rocky Mountain and adjacent regions. It rests upon the Montana marine group. The only upper limit now definitely known is a stratigraphic break, above which comes, normally, some formation of the Shoshone group. Should a series of sediments be discovered connecting in unbroken sequence the Laramie and the lowest recognizable Shoshone forma- tions, the boundary between them must be established at least for that section, on the basis of facts which cannot be anticipated. By this definition the Laramie is inseparably connected with the great Cretaceous section below it, as in the conceptions of King, Hayden, and perhaps all of the earlier observers. It was presumably coextensive with the Montana beds and represents conditions which had existed locally during the Judith River epoch in Montana. While the Laramie was originally so widespread, its distribution as now preserved, is a matter of much uncertainty. Veatch reports 6500 feet of Laramie strata below the horizon marking the break at the base of the Carbon beds and no doubt other developments of the Laramie occur in Wyoming, though recent investigations make it necessary to refer many supposed Laramie exposures of that State to the Shoshone group, and to study carefully all sections before making assignments. The Laramie is exhibited in its typical relations in the Denver basin and northward along the base of the Front Range. It there rests conformably on the Fox Hills (Eldridge) and is succeeded by a stratigraphic break representing a long period of erosion, as dem- onstrated by the materials of the basal conglomerate of the Arapahoe formation. The remaining Laramie of the Denver basin varies from 600 to 1200 feet in thickness. The greater part of the known Laramie flora has been found in this district in the coal measures near the base of the formation. Further discussion of Laramie distribution seems inadvisable at this time for the reason that the information gathered by many parties of the U. S. Geological Survey engaged in studying the coal measures of the Rocky Mountain region, under direction of M. R. Campbell, cannot as yet be utilized. 38 CROSS The Shoshone group may be defined as embracing the lacustrine, fluviatile or terrestrial deposits, composed of detritus from the rising land area of the Rocky Mountain province, formed between the Laramie and Fort Union epochs. The group embraces beds which have been commonly referred to as "Post-Laramie" or "Upper Laramie" in the literature of recent years. Dana, in his Manual of Geology, used the alternative terms "Upper Laramie" and " Denver Group, " the latter being an inadmissible and confusing application of Denver in a second, comprehensive sense. The name " Post- Laramie" was proposed by me, as already pointed out, for the formations now included in the Shoshone Group. Among the formations which may now be specified, on published data, as belonging to the Shoshone group are the Arapahoe, Denver, Middle Park, and Animas beds of Colorado; the Carbon and Evan- ston beds of Wyoming; the Livingston beds of Montana. The Converse County "Ceratops beds" of Wyoming and the Hell Creek beds of Montana belong to this group, according to Barnum Brown, ^* who calls them "Post-Laramie." Many of the formations of the Shoshone Group appear to have been quite restricted in area. At least, they are now isolated and correlation is as yet rather tentative. It is not yet known that the Arapahoe beds of the Denver Basin, or any other particular deposit, represent the earliest of these local formations. There is reason to correlate, more or less closely, the Denver, Middle Park, Animas, and Livingston beds, which consist largely of similar volcanic debris and contain similar floras or faunas. But it cannot be assumed that volcanic eruption was coextensive with the uplift of this period nor that it did not begin during the Laramie; hence it is probable that formations consisting of non-volcanic material were deposited in some districts during the Denver epoch, and there may be volcanic material in some Laramie strata. It is plain that before complete correlations of permanent value can be made among the Shoshone formations the stratigraphic relations must be thoroughly examined, the fauna and flora of each local deposit must be carefully studied in the light of stratigraphic evidence, and all elements brought into harmony. Such investigation will necessarily take many years. " The Hell Creek Beds of the Upper Cretaceous of Montana. Bull. Ainer. Mus. Nat. Hist., Vol. XXITT, 1907, pp. 823-845. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 39 Except for the brackish water forms of the Laramie fossils have not been cited in this discussion as characterizing the Laramie or Shoshone beds, although it seems certain that it will be possible, eventually to discriminate between them quite sharply in most cases by means of their fossil remains. It has always been an unfortunate feature in discussions of the Laramie question that there has been so little positive knowledge and so much assumption as to the true stratigraphic relations of fossil-bearing beds. At this time, when so many investigations are in progress, yielding abundant new material and revealing errors or deficiencies in earlier work, it seems best for the stratigrapher to regard the true significance of paleontologic data as imperfectly established, and I prefer to place the responsi- bility for characterizing the faunas and floras of the Laramie or Shoshone beds with the paleontologists. A few general statements may be made in illustration of the present situation. Fossil plants are widely distributed in all formations concerned. Without reviewing the early literature of the so-called "Laramie flora, " which was made to include plants actually occurring at vari- ous horizons between the Dakota and Miocene, it may be recalled that Ward's Synopsis of the Laramie Flora, ^^ published in 1885, was characterized by Newberry as "really and only an important contribution to our knowledge of the Fort Union flora."^^ Newberry further maintained that the Laramie (including beds here called the Shoshone Group) and Fort Union floras were totally distinct. Knowl- ton has endorsed Newberry's view as to the essentially distinctive character of the Fort Union flora, '^ and, through study of many new collections, partly gathered in person, he is now in a position to affirm that opinion. By Mr. Knowlton's courtesy I am permitted to refer here to the fact, shown by recent collections, that the Fort Union flora ranges downward into the "Hell Creek beds." The most important collection on which this statement rests was made by Barnum Brown, the plants being in direct association with the "Ward, Lester F. : Synopsis of the Flora of the Laramie Group, U. S. Geol. Survey, 6 Ann. Rep., 188, pp. 399-557- ^« Newberry, J. S. : The Laramie Group, Bull. Geol. Soc. Am., Vol. 1, 1890, P- 525- "Knowlton, F. H.: Notes on a Few Fossil Plants from the Fort Union Group of Montana, etc., Proc. U. S. Nat. Mus., Vol. XVI, 1893, pp. 33-3(>- 40 CROSS huge dinosaurs. The flora of the Fort Union group now known to Mr. Knowlton embraces about 500 species. The discovery of the Fort Union flora in association with the vertebrate fauna of the Hell Creek beds raises several highly impor- tant questions, such as the relative value of plants and dinosaurs in determining the age of the Fort Union. A discussion of these ques- tions and a comparison of the Fort Union, Shoshone and Laramie floras will be given in a forthcoming paper by Mr. Knowlton. The Shoshone flora known to Mr. Knowlton now embraces about 300 speecies, 200 of which occur in the Arapahoe and Denver beds. The known flora of undisputed Laramie beds of Colorado contains 80 species. It is sufi&cient to say here that the Shoshone flora is distinctly intermediate in character between those of the Laramie and Fort Union and is easily distinguished from either, according to Mr. Knowlton. The invertebrate fauna of the Laramie contains both fresh and brackish water forms. The latter belong to Osirea, Corbula, Cor- bicula, Anomia, etc., and are identical with or nearly related to forms also known in brackish water deposits of the Montana group, as at Coalville, Utah, and Point of Rocks, Wyoming, and in Judith River beds. Such forms are rarely found, I believe, in the Shoshone beds; so rarely that there must arise in each case a query as to the signifi- cance of the presence of these forms. Does it necessarily imply a subsidence permitting a recurrence of brackish water conditions, or may the forms in question have adapted themselves to fresh water environment ? The intimate association of brackish and fresh water forms above the Laramie may well raise the latter question in any case. The remarkable vertebrate fauna occurring in beds referred to the Laramie in a general way by paleontologists has until recently been a hindrance rather than a help to the stratigrapher engaged in studying the geologic development of the Rocky Mountain region at the close of the Mesozoic era. During the years in which Marsh was engaged in describing in rapid succession the wonderful horned dinosaurs and associated forms "from the Laramie of Colorado," Wyoming, or Montana he came to speak of the "Ceratops beds" and the "Ceratops fauna" collectively, minimizing almost to the degree of ignoring, the importance of stratigraphic and other evidence showing that the vertebrates he grouped together so closely came THE LARAMIE FORMATION AND THE SHOSHONE GROUP 4 1 from several formations which were clearly not synchronous deposits. It would appear to be self-evident that the vertebrate life must have undergone marked evolution during the series of orogcnic disturbances characterizing the Laramie and Shoshone epochs. Hatcher, who took up the study of the Ceratops fauna, after Marsh, realized the importance of careful stratigraphic examination, and before his lamented death laid the foundation for the phylogenetic study of the "Laramie" vertebrates which alone can make them of diagnostic value in determining stratigraphic relations. The inves- tigation of the Judith River beds of Montana by Hatcher and Stan- ton in 1903 demonstrated that they belong in the Montana Group. The "Ceratops fauna" was thus carried below the Laramie, and Hatcher, instituting such comparisons as were possible, announced the general conclusion as regards the vertebrates, that In every case where any group of the fauna has been studied from suflS- cient material, it has been found to be represented by distinctly older and more primitive types than the related forms from the Laramie.^" In harmony with this view are the opinions of Stanton concerning the invertebrate fauna and of Knowlton on the flora, to be found in the same publication. Vertebrate remains have been obtained in the Laramie beds (as here defined) in but very few locahties. The most important fossil to be referred to here is the Agathaumas silvestris Cope from Black Buttes, Wyoming. This form is represented by skeletal parts and cannot therefore be satisfactorily compared with most of the other Ceratops forms, which are known only from cranial parts. The recent review of the phylogeny of the Ceratopsia by LulP^ places Agathaumas silvestris by itself as a form much older than Tricera- tops of the Converse County beds. This view agrees with the refer- ence of the Black Buttes beds to the Laramie by Stanton on the evidence of brackish water forms believed to occur shortly above the Agathaumas horizon, but conflicts in some measure with the close 2°T. W. Stanton and J. B. Hatcher: Geology and Paleontology of the Judith River Beds with a Chapter on the Fossil Plants by F. H. Knowlton. U. S. Geol. Survey, Bull. 257, 1905, p. 103. " The Ceratopsia, by J. B. Hatcher, Mon. XLIX, U. S.Geol. Surv., edited and completed by R. S. Lull, 1907, Part II by Lull, Phyolgeny, etc., p. 161. 42 CROSS correlation of Black Buttes and Converse County beds suggested by the similarity of their freshwater faunas and the floras. ^^ Barnum Brown correlates the Converse County and Hell Creek beds on the basis of unconformable relations to marine Cretaceous, lithologic character, vertebrate and invertebrate fauna, flora, and position relative to Fort Union beds. He also calls them " Post- Laramie" and extends the correlation to the Denver beds because of the Triceratops fauna in the latter.-^ The Triceratops is described by Hatcher and Lull as the most highly differentiated form of the horned dinosaurs known. It was apparently the last of the race. The mammals associated with the Triceratops fauna in Converse County, whose Jurassic affinities were so strongly emphasized by Marsh, must manifestly be studied as to their relations to the forms of the Torrejon and Puerco, before their significance can be determined. The facts recited seem to me to indicate clearly that when the distribution of fossils of all kinds in the Laramie and several Shoshone formations is more completely known and the phylogeny worked out, it will be found that each formation has its distinctive fauna and flora which can be utilized with certainty in correlating newly dis- covered deposits of obscure relations. In the preceding discussion I have avoided the question as to the age of the Shoshone beds, whether Cretaceous or Eocene. I desire now to urge their reference to the Eocene. The Denver beds were originally referred by me to the Eocene, but the great weight attached to the Mesozoic affinities of the vertebrate fauna by paleontologists led to a tentative acquiescence in the assignment of the Arapahoe and Denver formations to the Cretaceous, in the Denver monograph. In that volume I reviewed various aspects of the question and can add but little to what was there said. The main point seems to be that the Laramie and Shoshone beds belong to a transition series between the Cretaceous and Eocene and that whatever break occurs between any two formations is possibly bridged over by deposits of some other locality. The Laramie is related to the Judith River and other brackish water formations of the IMontana Cretaceous, "'Stanton and Knowlton, Bull. Geol. Soc. Amer., Vol. VII, 1S98. ^^ The Hell Creek Beds of the Upper Cretaceous of Montana. Bull. Amer. Mus. Nat. Hist., Vol. XXIII, 1907, pp. 823-845. THE LARAMIE FORMATION AND THE SHOSHONE GROUP 43 the Shoshone to the great succession of Tertiary local deposits. The retreat of marine waters and the decided uplift of a large continental area marked the appropriate boundary between Cretaceous and Eocene from the stratigraphic side. The plant life seems to show no special reason for drawing the line at any other horizon. The argument most effectively used for including the Shoshone beds in the Cretaceous has come from the vertebrate remains of that group. The Mesozoic affinities of that fauna are said to be so pronounced that paleontologists have been unwilling to consider the proposition of referring beds containing that fauna to the Eocene. The intimate relations of the fauna to its ancestors are not surprising and as a large part of the fauna became extinct in Shoshone time the absence of a comparable line of descendants is easily compre- hended, but the stratigrapher naturally raises the question— Why is the extinction of the huge vertebrates of the Shoshone epoch to be considered as determining the Mesozoic-Cenozoic line? A quarter of a century ago it was the common belief of stratigraphers and paleontologists that the "great Rocky Mountain revolution" so changed conditions of environment affecting the living forms of the time that the Mesozoic vertebrate fauna in particular could no longer exist. Little was then known of this revolution and yet an almost catastrophic influence of the uplift upon life was assumed as natural. Now that the change from marine to permanent continental condi- tions is known to have been marked by oscillations and the verte- brate fauna is shown to have survived the very changes once assumed to have been fatal to it the cause of extinction becomes a mystery, yet is held to have the same great significance formerly assigned to it. This conclusion is surely open to question. The little mammals fouad in the Converse County beds have been charged with exterminating the huge dinosaurs, but Lull rejects this hypothesis and reverts to orogenic movement as the cause.^* But there is at present no available evidence to show a movement of any unusual importance between the Denver and Fort Union epochs. It has been suggested by Peale that the extinction of vertebrate faunas in the Miocene and Pliocene periods was due to widespread showers of volcanic ashes and the same cause has recently been 2* The Ceratopsia, U. S. Geol. Surv. Mon. XLIX, p. 44 CROSS brought forward by Scott to account for the extinction of the Miocene fauna of Patagonia. If the disappearance of the Ceratopsia and associated reptilian forms may possibly have been due to the great volcanic outbursts of the Shoshone epoch, it is nevertheless true that no ash beds have been found in which their remains occur in per- fect condition, as is the case in the other instances referred to. The Denver beds are for the most part not ash beds. It is of course possible that such beds were eroded before the Fort Union epoch, but this explanation is certainly as yet merely a speculation. Where a gradual change takes place from one set of conditions to another various groups of animals and plants of the earlier period will adjust themselves with varying rapidity and success to the con- ditions of the later time, and some may disappear through the lack of the power of adjustment. The forms which survive longest are manifestly not those which mark the change in conditions. The vertebrate fauna can logically be used to limit the Cretaceous of the Rocky Mountain province only on the basis of a time correlation through those forms with some well established section elsewhere. As far as I know there is no such section on the American continent. Certain marked resemblances have been pointed out between the Ceratops fauna and that of the Gosau formation in Austria, which, after much discussion has been referred to the uppermost Creta- ceous. But it is of much importance in citing this resemblance to know whether it is the Triceratops fauna of the Shoshone group, the forms of the Judith River Cretaceous, or possibly a more primitive fauna, which is most closely related to that of the Gosau beds. The question as to where the Cretaceous- Eocene line shall be drawn in the Rocky Mountain province is one to be decided on broad grounds and in closing I wish to quote the philosophical views of Dr. C. A. White, a biologist, paleontologist, and strati- grapher, expressed in regard to this very question nearly 20 years ago, after discussing the evidence of the Arapahoe and Denver beds. Referring to precise determinations of the age of formations solely on fossil evidence. Dr. White remarks: Those paleontologists who make this unwarranted application of their science to systematic geology, all use the scheme of classification that has been established for Europe, and use it as if it were of infallible a})i)lication to all other parts of the world, and also as if it were already absolutely THE LARAMIE FORMATION AND THE SHOSHONE GROUP 45 perfected for that continent. While I have no inclination to question the general accuracy of the European scheme of classification for that continent, I do not hesitate to express the opinion that it is not of infallible application to other parts of the world, except as to its larger divisions, and that even in this respect it will need modification. That is, I hold that investigation of the formations which are found upon any given continent or great divi- sion of the earth's surface ought to be prosecuted, first, with relation to one another, and second, with reference to their ultimate, not immediate, cor- relation with those of other continents or divisions. ^° " Vice-Presidential address, The North American , Mesozoic, Proc. A.A.A.S., Vol. XXXVIII, 1889, pp. 225-226. PROCEEDINGS OF THE WASHINGTON ACADEMY OF SCIENCES Vol. XI, No. 2, pp. 47-178 April 28, 1909 STUDIES IN THE AMERICAN BUPRESTID^. By Thos. L. Casey. It would seem that but few sections of the Coleoptera have been so neglected, or so superficially investigated with a view to scientifically defining and grouping the genera, as the Buprestidae. The conspicu- ous and brilliant metallic coloration prevailing among the species, doubtless led the early authors to define the majority of them from their salient external characters alone, and generally in few words, so that the determinative literature is inadequate and frequently mis- leading. These remarks apply, however, more especially to the first few groups of genera as they occur in America, for some careful and undoubtedly useful work has been accomplished in subsequent parts of our series, for example in the genera Chrysobothris, AcmcBodera and Agrilus by Dr. G. H. Horn and Mr. H. C. Fall. Alluding to the neglect during the past fifty years of that part of the family which forms the subject of the present essay, it will suffice to state that only a comparatively small proportion of the specific and subspecific forms in our cabinets have been defined, that the genus Chalcophora has been constituted in our lists during all this time of two notably distinct genera, that the three species assigned to Hip- pomelas belong in reality to three different genera, two of which are not at all closely allied, and, finally, that Buprestis is separated from the earlier genera by Dicerca and Poecilonota, which differ conspicu- ously in antennal structure as will appear below. This general lack of interest in the taxonomic treatment of the family is apparently due, in some measure at least, to actual absence of structural plasticity, the entire under surface of the body being unusually fixed and con- Proc. W^sh. Acad. Sci., April, 1909. 47 48 CASEY stant throughout, so that even Lacordaire, generally so happy in his choice of primary characters, has, seemingly in despair, seized upon a few antennal features for major group division, without testing them with sufficient thoroughness; the result is that the grouping suggested by that author, especially that founded upon variations in the ridge above the antennal foveas and in the pores and fossae of the outer antennal joints, is not satisfactory, as indeed remarked by LeConte in his monograph of the family. At the same time the group char- acters adopted by LeConte are also valueless to a great degree, for it is not apparent that the front is narrowed by antennal insertion to any greater degree in Chrysohothris than in Gyascutus, and the nature of the meso-metasternal suture, although very important, completely fails as a group character in the sense intended by that author, while the differences in prosternal conformation are too feeble, inconstant and illusory to serve any useful purpose. It is perhaps also due in part to this lack of evident organic structural diversity, that early mis- takes in generic assignment were so frequently made by systematists; but it is difficult to comprehend, even under this assumption, the original referring of what we now know as Gyascutus planicosta and ohliteratus, as well as many exotic species, to the genus Chalcophora. It has been impracticable for me to survey the entire family with sufficient thoroughness to definitely pronounce upon primary group division, if there be any worthy of adoption, and the genera here reviewed are therefore arranged in close succession in a single state- ment and without tribal delimitation. The principal generic charac- ters assumed relate almost wholly to antennal and elytral structure, but, for more closely related genera, diverse characters of the front, pronotum and tarsi have proved to be of more or less utility. There are two sufficiently distinct types of antennal structure among the fjenera having smoothly anchylosed sterna, and, if found to be of gen- eral application in the family at large, there is no reason why they should not form the basis of major group or tribal division. In the first group we might suggest Hippomelas and allied genera, having a terminal antennal process, as one tribal group, Psilopiera, Spinthoptera and probably others now confounded with Psilopiera, as another, Chalcophora and related forms as a third and Buprestis with its im- mediate allies as still another, and, in the second division, which is probably by far the larger, Dicerca, with related genera, should be STUDIES IN THE AMERICAN BUPRESTIDiE 49 considered a special tribe based upon the position of the antennal fossae and distinguished from others, such as Chrysobothris, by the form of the elytral bases, degree of separation of the eyes and other characters. Many genera were unnecessarily rejected by Lacordaire, presumably because of deficiency in structural peculiarities affecting those special organs usually looked to elsewhere, or in diversification of the sclerites of the under surface, overlooking the fact, of which we here have sufficient proof, that our criteria for generic definition should vary to fit the conditions of the particular section of the Cole- opterous series that we may have under consideration. In this case elytral and antennal structure become of paramount importance from the generic viewpoint, although in many other sections of the order they may be wholly unimportant. Messrs. Warren Knaus, H. F. Wickham and Chas. W. Leng have generously permitted me to acquire duplicate material from their collections, and this, together with a considerable series in my own cabinet collected by Dunn, Levette and others, gives me a very fair representation of the species. I also have to thank Mr. Champion for a transcription of Ollivier's diagnosis of Buprestis viridula. In order to study the laws of intraspecific variation I obtained extensive series collected by Mr. A. H. Manee, at Southern Pines, North Caro- lina, consisting of about sixty-five specimens of Buprestis ornata and some twenty-five examples each of B. apricans and Dicerca obscura. These have been of some utility in several efforts to decide whether or not two closely allied uniques might represent different species, for the laws of variation applying to a few species are very apt to apply to all those in the same group of genera. The secondary sexual char- acters at the apex of the abdomen are remarkably inconstant within rather broad limits and cannot be employed, as a rule, in defining allied species, and punctuation and general sculpture are also variable in some parts of the series to a rather unusual degree; but the forma- tion of the elytral apices is usually rather constant. So it required a large amount of study to become sufficiently familiar with the subject to attempt the definite separation of specific and subspecific forms. Probably I have made some mistakes which future knowledge will rectify.^ 'This paper is the gradual outgrowth of an original intention on the part of the writer, to describe, merely in an isolated manner, a few species of Gyascutus. 50 CASEY The genera here considered may be defined as follows : — Mesosternum and metasternum anchylosed, the suture obsolete or very fine 2 Mesosternum separated from the metasternum by a distinct cleft 14 2 — Outer antennal joints more or less obliquely truncate beneath, always more rapidly or abruptly narrowed basally, and usually with sensitive setigerous pores in addition to the fossae, the latter in a few cases wholly obsolete 3 Outer antennal joints enlarged within apically, or more typically serriform, not more rapidly or abruptly narrowed basally, usually without aggre- gated sensitive pores but always having an intero-terminal sensory fossa; antennae shorter and more compact as a rule 10 3 — Last antennal joint with a terminal process, wanting apparently only in the male of Hippomelas; sensory fossae terminal, diversely developed and sometimes obsolete; pronotum never sulcate. Sonoran faunal regions 4 Last antennal joint without trace of terminal process in either sex 6 4 — Last antennal joint of the male without evident terminal process; body moderately stout to narrow and subcuneate, convex, more feebly and evenly sculptured, the sides of the prothorax more evenly converging from base to apex and frequently nearly straight; tarsi variable. [Type Buprestis saginata Mann .] Hippomelas Last antennal joint having a distinct terminal process in both sexes; sculp- ture of the integuments deeper and more irregular 5 5 — Pronotum with the basal bead never entire but always interrupted medially, where it joins the raised median smooth line; tarsi more slender and elongate, the basal joint of the posterior almost always longer than the next two combined; sculpture moderately diversified. [Type ChalcopJwra planicosla Lee] Gyasciitus Pronotum with the basal bead strong, even and entire, not obliterated at the middle; tarsi shorter and thicker as a rule though variable, the basal joint of the posterior not as long as the next two combined; sculpture rough though more uniform, that of the pronotum almost as in Spinthoplera. [Type Chalcophora calata Lee] Stictocera 6 — Outer antennal joints with lateral as well as apical sensory fossae. Body more or less parallel, the apical part of the elytra regularly ogival as a rule and narrowly truncate and bispinulose at the aj)ices; prono- tum wholly devoid of basal beading, not sulcate though sometimes in part feebly, longitudinally impressed medially; elytra with impressed series of coarse punctures, and, in addition, scattered finely punctulate and puberulent shallow areolae, which are usually bright metallic in coloration; prosternum bisulcate, transversely truncate anteriorly; basal segment of the abdomen feebly ilattened to deeply sulcate; tarsi thick, the first joint of the posterior but little longer than the second. Sonoran and Neotropical regions. [Type Psiloptera valcns Lee.]. Spinthoptera STUDIES IN THE AMERICAN BUPRESTIDvE 5I Outer antennal joints with the sensory fossae single and intero-terminal, very rarely obsolete 7 7 — Elytra very irregularly sculptured; pronotum always sulcate; epistoma unusually narrowly and deeply emarginate 8 Elytra with regular sculpture, never having the large, uneven and interrupt- ing depressions of the preceding group; pronotum generally without a median sulcus though sometimes impressed along the median line; epistoma more broadly and feebly sinuate as usual; legs slender. Holarctic faunal region 9 8 — Pronotum bisulcate, the median line more or less evidently costuliform; elytra with deep irregular depressions, the side margins never more than feebly serrulate posteriorly, the apex generally obtusely angulate ; antennae with the setigerous pores scattered and perforate, the sensory fossa obsolete; labrum corneous; mentum very transverse, truncate, wholly corneous; presternum bisulcate. Holarctic regions. [Type Buprestis nmriana Linn.] Chalcophora Pronotum unisulcate; antennae rather thick, the outer joints with close- set impressed setigerous pores and with well developed intero-apical sensory fossae, the latter irregular in form; labrum coriaceous; mentum less transverse, trapezoidal, corneous in more than basal half, the apical part coriaceous; elytra irregularly punctured, very unevenly costulate and with very large and shallow interrupting areolae, the lateral edges strongly serrate posteriorly, the apex not truncate; presternum and first ventral impressed along the median line. Nearctic regions. [Type Buprestis campestris Say] Texania 9 — Scutellum large, the remaining characters nearly as in Buprestis, the elytra regularly striate. Europe. [Type Buprestis austriaca Linn.]. * Eurythyrea Scutellum small; elytra striate; labrum generally wholly coriaceous and pale in color but becoming corneous and metallic basally in some aberrant species; elytral apices rounded or truncate, sometimes bispin- ulose, the sides not serrulate posteriorly; prosternum convex or llat- tened along the middle; first ventral convex to deeply sulcate; basal joint of the hind tarsi more or less considerably elongate. Holarctic regions. [Type B. octoguttata Linn.] Buprestis Elytra not striate but regularly costate, with densely cribrate and broadly concave intervales, without trace of serial punctures; labrum always corneous and metallic basally; prosternum impressed and punctured along the middle; elytra rounded or truncate at tip. Nearctic re- gions. [Type Ancylocheira lauta Lee] Cypriacis Elytra densely and subevenly cribrate, without striae, but having regular series of large perforate punctures; remaining characters nearly as in Cypriacis. Atlantic Nearctic regions. [Type Buprestis decora Fabr.] Stereosa 10 — Hind coxae strongly dilated within as usual, their hind margin thence strongly oblique to the sides; antennal fossae antero-intemal, small and 52 CASEY rounded; eyes with their inner margins oblique as usual; tarsi longer; body not cuneiform 1 1 Hind coxae but slightly dilated within, their posterior edge thence transverse to the sides of the body; eyes parallel within; antero-terminal fossae of the outer antennal joints large and irregular, the joints also with a small lateral fossa in addition; body subcuneate in form 13 n — Pronotum more closely punctate or depressed along the median line; elytra roughly sculptured, more or less prolonged at tip, not or scarcely serrate at the sides posteriorly; antennae inserted in larger depressions under more or less pronounced oblique elevations of the front. Hol- arctic regions. [Type Buprestis cenea Linn.] Dicerca Pronotum subelevated and impunctate along the median line; elytra less roughly sculptured and usually with more distinct striation 12 1 2 — Antennae inserted under oblique prominences somewhat as in Dicerca; elytra more or less feebly produced at tip, not or scarcely serrulate at the sides, the apices truncate or sinuate and generally bispinulose; surface with sombre coloration. Holarctic regions. [Type Buprestis variolosa Payk.] Poecilonota Antennas inserted in exposed foramina, the supra-antennal ridges obsolete; elytra not produced at tip, feebly serrulate at the sides posteriorly, the apices more or less truncate and multispiculose; integuments with vivid metallic coloration. Europe. [Type Buprestis festiva'Lmn.]. *Lampra 13 — Antennae inserted in small frontal foramina, very short and compact; pronotum feebly impressed medially toward base, otherwise almost evenly convex ; scutellum small; elytra subevenly cribrate and with very faint subconvex smoother Hnes, the sides serrate posteriorly and at the narrowly rounded apex, which is not at all produced; tarsi very short. Sonoran regions. [Type Halecia gentilis Horn] Anataxis 14 — Scutellum small as usual but distinct and normal 15 Scutellum obsolete; elytra without striation 16 15 — Antennae inserted in small foramina under very small, feeble and inconspicuous ridges, short, compact, the joints nearly as in Dicerca but more convex and with dense sensitive pores beneath, the sensory fossae intero-terminal, distinct and circular, the last joint with a strong terminal process as in Gyascutus; body small, subcylindric, the eyes widely separated, with their inner margins but slightly converging up ward; prothorax subparallel and broadly rounded at the sides, faintly impressed along a part of the median Hne; legs slender; tarsi short; hind coxaj dilated internally as usual; first ventral suture distinct, rectihnear throughout; elytra feebly serrulate at the sides and finely striato-punctate. California. [Type A'^. cupreofusca n. sp.] Nanularia Antennae inserted in very small and widely separated foramina, which have their upper edges horizontally acute but not ridged, very slender, the outer joints elongate and much as in Buprestis, sparsely punctate, more closelv and subrugulosely beneath, with very small rounded intero- STUDIES IN THE AMERICAN BUPRESTID^ 53 terminal sensory fossa; labrum well developed, sinuato-truncate; epistoma unusually deeply and not broadly emarginate; body small, slender; eyes with the inner margins distinctly converging upward as usual; prothorax feebly trapezoidal with nearly straight sides, trisul- cate; elytra without trace of serrulation at the sides, sinuate and bi- denticulate at the apices, obscurely and imperfectly striate and with uneven and undefined subimpressed areolae; prosternum not sulcate; first ventral convex, the first suture fine but distinct, straight; legs long and very slender. Atlantic regions. [Type Buprestis elongata Lap.-Gory] Cinyra 16 — Antennae inserted in small exposed foramina, not within a cavity and without superior ridge though having a small polished and sculpture- less spot adjoining each above and becoming more prominent along the upper margin of the foramina, their outer joints strongly serrate and triangular, with their lower margins very oblique and more converging basally, having dense asperate sensitive pores beneath anteriorly and a large intero-apical sensory fossa, the last joint narrow and elongate- oval; eyes widely separated, their inner margins sUghtly converging upward; epistomal margin narrowly, deeply sinuate medially; labrum as in Cypriacis, coriaceous but having basally two large dense and metallic areas which are very narrowly separated medially; maxillary palpi very slender; mentum sinuate at apex, corneous and densely cribrate ; pronotum uneven, with a short compressed and prominent ridge at each side behind the middle of the lateral margin and a small deep cavity before the middle of the base; scutellum wholly obsolete; elytra oblong, wider than the prothorax, cribrate, without trace of striae and very uneven, scarcely at all serrulate at the sides, the tips obtuse, round- ed; prosternum densely sculptured throughout, not impressed,^ the abruptly narrowed apex not attaining the metasternum, the side pieces of the mesosternum separated, however, by a deep cleft; basal segment of the abdomen apparently free or nearly so, the first suture similar to the others and rectiUnear; legs slender, the basal joint of the hind tarsi greatly elongated; hind coxae strongly oblique as usual. Pacific coast to Gulf of Mexico. [Type T. hlondeli Mars.] Trachykele The usual position of Dicerca, interpolated between Hippomelas and Buprestis, dates apparently from the work of Lacordaire, who drew attention to the similarity in antennal insertion and form of the supra- antennal ridges to those of Psiloptera; but the marked dissimilarity of the antennae themselves seems to the writer a far more important consideration and necessitates the removal of Dicerca and Poecilonota from the vicinity of Hippomelas, Psiloptera and Buprestis, to a closer association with Chrysohothris. The European Eurythyrea Sol., and Lampra Spin., are introduced above for comparison and greater com- pleteness; they are distinguished by the prefixed asterisk. There can be but little question that Lampra ought to be considered a genus dis- 54 CASEY tinct from Poecilonota and not a subgenus. Anataxis, proposed for Halecia gentilis, is not closely related to Dicerca or Poscilonoia, but I can imagine no better place for it than that assigned in the table. It was placed in Agaocera, by Waterhouse, but on examining the figure given by Laporte and Gory for Atithaxia gigas, the type of AgcBocera, I cannot perceive the least serration of the sides of the elytra and, besides, the pronotal and elytral structure is so different that I venture to propose for gentilis a separate genus. This species is entered twice in the Kerremans Catalogue, once under AgcBocera and again under Halecia. The genus Trachykele, of Marseul, is very aberrant, not only in general facies and absence of scutellum, but in the virtually free basal segment of the abdomen. It is quite prob- able that Buprestis lecontei Lap.-Gory, constitutes a genus different from that founded upon such species as hlondeli and opulenta. The delimitation of species in this group of genera, and indeed throughout the BuprestidcT, is rather more difficult and uncertain than usual, because of erratic variation in many of the elements generally relied upon for the definition of species. Kerremans alludes to this in the introduction to his general catalogue of the family, and intimates that there may possibly be no such reality in nature as species or genera, or even higher groups, and that it may be all a matter of individual opinion. The expression, "opinionative spe- cies," is frequently employed in alluding to those which are main- tained or reduced from time to time through the idiosyncrasies of writers, though rather meaningless after all, because of the imprac- ticability of standardizing human intelligence; but, on the strength of his expressed doubts and apprehensions, the author quoted has had recourse to a wholesale system of " lumping, " which appears to me not fully justified. Because of deficiency in material I have not always been able to make up my mind definitely as to species, having had to rely more on general habitus than anything else in a number of cases. Those forms which are consequently to some extent doubtful to me as species are inscribed as subspecies in the various tables, but, because of the undesirability of trinomials, these subspecies should be quoted under two names only when referred to verbally or in text; they would not have been named at all if there were not a very strong prob- ability of true specific value in every case. In some parts of the series I have united and separated forms many times before coming to a STUDIES IN THE AMERICAN BUPRESTID^ 55 conclusion, final as far as apparently permissible from the incomplete material at hand. Kirby's species were all collected from three to ten hundred miles north of our northern boundaries and some of them are unquestionably different from our own; the attempt to force some of our forms to bear the Kirbyan names has given rise to several mis- takes. The Buprestis viridula, of Ollivier, described from Carolina, is a Philippine species of Dicercomorpha. Gyascutus californicus, of Horn, is assumed to form part of the genus Nanularia, though in that species the last antennal joint is specially described as obtuse at apex, so that this reference is somewhat hypothetical. If califor- nicus is not assignable to Nanularia, however, I can suggest no other place for it, as, apparently, it can be neither a Gyascutus nor a Hip- pomelas. Hippomelas Lap.-Gory. The species of this genus are more elongate and subcylindric as a rule than in Gyascutus and Stictocera, and of much smoother and more feebly sculptured surface; they, however, occasionally become rather stout in form, as in the type, which was named saginatus by Mannerheim; this species is not represented in my collection at pres- ent. The coloration is usually black and frequently without trace of metallic lustre, but is occasionally bright and metallic, particularly in mexicanus and related species of southern Mexico, and also in the second subgenus, defined below. The head and eyes are large, the frontal surface roughly sculptured as usual, descending at the sides in a more punctate and pubescent slope or depression to the inner margin of the eyes, the epistoma broadly, feebly sinuate, the antennae moder- ately long and compressed, the outer joints with dense sensitive punc- tures, except above, and having an irregular and somewhat inconstant or vestigial apical sensory fossa near the margin of the articular for- amen, and they are inserted under strong oblique frontal ridges. The prosternum is not impressed though feebly flattened, the meso- sternum wholly divided, the meso-metasternal suture fine and some- times barely traceable, the first ventral convex medially and the legs variable subgenerically. There are two subgenera as follows: — Tarsi shorter and thicker, the basal joint never as long as the two following combined; abdomen with the basal segment shorter, behind the coxae more or less evidently shorter than the second, the first suture sinuate medially as usual though almost obliterated Hippomelas 56 CASEY Tarsi longer and more slender, the basal joint fully as long as the next two or longer, as in Gyascutus; abdomen with the basal segment a little longer, the first suture similar but better marked Prasinalia The latter of these subgenera seems to be peculiar to the fauna of Arizona and the adjacent parts of California, where it replaces Hippo- melas, which is widely diffused from our southern boundaries to southern Mexico. Hippomelas in sp. Legs very thick, the tarsi unusually short, the basal joint of the posterior but little longer than the second. Body moderately narrow, convex, subcylindric, cupreous in lustre, the elytra more aeneous, the epipleurae and sides of the under surface becoming greenish; head ( c?) much more depressed throughout the frontal surface and vertex than in any other species, feebly concave, hairy, more finely and uniformly sculptured and green from the occiput to the strong antennal ridges; outer antennal joints parallel; prothorax two-fifths wider than long, moderately trape- zoidal, the sides straight; apex feebly arcuate except laterally, the base broadly bisinuate; surface moderately strongly, subevenly and sparsely punctate, more closely laterally and with feebly marked and more convex vacant spaces throughout, the lateral margin green; scutellum small, transverse; elytra slightly wider than the prothorax and three and one-half times as long, the sides feebly converging from the humeri to about apical third, thence rounding and with long, ver}^ feeble ser- rules to the apices, which are narrowly and deeply emarginate and briefly bidentate; surface even, with barely impressed striae of fine close punctures, becoming larger and closely confused at the sides; intervals scarcely punctate, somewhat creased; under surface sparsely punctate, polished, bright coppery, the legs and lateral parts unusually pubescent. Length 23.0 mm. ; width 7.5 mm. Mexico (Guerrero) . *brevipes n. sp Legs less thick, the tarsi less abbreviated, the basal joint of the posterior almost as long as the next two combined as a rule 2 2 — Body stouter in form, always black or blackish and seldom with notice- able metallic lustre. Sonoran regions 3 Body narrower, decidedly slender though convex and subcylindric, bright metallic in lustre; third antennal joint much shorter, but little longer than the fourth. Southern Mexico 6 3 — Elytra with very feeble aeneous or greenish-metallic lustre 4 Elytra deep black or piceous-black, without trace of metallic lustre 5 4 — Form elongate-oval, convex, stout, moderately shining, the hairs of the elytra very short and sparse, dense and longer as usual at the finely punctate basal margin, also unequally on the head, the legs very sparsely pubescent; head large, with large irregular elevated smooth areas which are separated by aeneous depressions, the latter densely and rather finely punctate; antennae moderate, the third joint unusually elongate; pro- STUDIES IN THE AMERICAN BUPRESTID^ 57 thorax three-fourths wider than long, the sides broadly rounded, more strongly converging anteriorly, the basal angles somewhat prominent; apical margin broadly arcuate except laterally, the base sinuate later- ally; surface rugose laterally, and, less distinctly, apically, rather finely, remotely punctate medially, the interspaces laterally with dense leneous punctures in vermiculate series; scutellum transversely oval; elytra not evidently wider than the prothorax, narrowed slightly to near apical third, the sides thence rounding and feebly serrulate to the feebly emarginate and bidenticulate tips; surface convex through- out, striato-punctate nearly as in brevipes; under surface shining, black and aeneous, the abdomen strongly, rather sparsely aeneo-punctate, convex, notably ascending posteriorly. Length 28.0-29.0 mm.; width io.c^io.3 mm. Western Texas grossus n. sp. A — Nearly similar to the preceding but shorter, the prothorax somewhat shorter, more strongly punctured, less narrowed anteriorly, the head relatively still larger; elytra shorter, more obtuse behind, the series of punctures scarcely at all impressed and the punctuation laterally less confused and scarcely more close-set than toward the suture; tibise shorter, more outwardly expanded at apex and notably more densely setulose throughout. Length 26.5 mm.; width 9.5 mm. Western Texas robustusn. subsp. Form narrower and more elongate, smaller in size, subcylindrically convex, shining, the head large, in great part aeneous, densely punctured and pubescent, the smooth irregular areas strongly elevated; antennae strongly compressed as usual, the third joint slightly less elongate; eyes large and prominent; prothorax deep black, but little more than one-half wider than long, the sides feebly converging, broadly, feebly and subevenly arcuate from base to apex, the latter feebly arcuate; base deeply sinuate laterally, the angles not at all everted; surface deeply vermicularly rugose, with the intervals aeneo-punctate, coarsely punctate apically, less coarsely and sparsely medially and basally; "scutellum tumid, transversely oval; elytra but little wider than the prothorax, formed nearly as in grossus and similarly sculptured, with widely spaced and distinct punctures in uneven single series along the intervals, greenish-metallic in lustre, the sides apically rather strongly and acutely serrulate, the apex approximately bidentate as usual; under surface in great part aeneous, the abdomen strongly, more closely and irregularly punctured, less strongly ascending posteriorly than in grossus; tibiae not densely but strongly albido-setulose. Length 25.0 mm. ; width 7.8 mm. Arizona (Tucson) serrulatus n. sp. Form elongate, convex, much smaller in size than grossus, blackish-aeneous in color, the pubescence sparse and extremely short; prothorax one-half wider than long, slightly narrowed anteriorly, the sides broadly rounded, the posterior angles acute; elytra finely striato-punctate, the intervals sparsely punctate. Length 18.8-25.0 mm. Texas (Laredo to Ring- gold Barracks). [=Buprestis sphenicus Lee] sphenicus Lee. 5— Body rather short, stout, the male narrower and with more cuneate elytra, deep black, slightly shining, subglabrous as usual; head moderately 58 CASEY large, less elevated between the eyes than in the preceding species, rugose and with greenish punctures, pubescent, more concave anteriorly in the male; eyes large, prominent; antennae extending nearly to the base of the prothorax (c?) or evidently shorter ( 9 ) ; prothorax trape- zoidal, three-fourths to four-fifths wider than long, the sides distinctly arcuate, becoming feebly sinuate basally, rugosely aeneo-punctate, smoother and sparsely punctate medially and toward base; scutellum less transverse and more quadrate than in the preceding species; elytra scarcely wider than the prothorax, of the usual form, scarcely at all ser- rulate at the sides behind, with the usual apical sinuation, the surface notably flattened toward apex, the punctured series more or less con- fused or irregular except toward apex, where they become deeper, excepting four pairs of striae on each elytron, which are regular through- out; under surface black and aenescent, the abdomen coarsely and rugosely but not densely punctate; legs moderate, the hind tibiae more gradually and distinctly thickened from base to apex in the male, the fifth ventral truncate in both sexes but with the marginal gutter deeper and more abrupt in the male. Length 23.0-27.0 mm.; width 8.3-10.0 mm. Southern Arizona planicauda n. sp. Body parallel and more convex, deep black, somewhat shining; head rela- tively rather small, nearly as in planicauda, the eyes large and prom- inent; third antennal joint a little more than twice as long as wide; prothorax feebly trapezoidal, three-fourths wider than long, the sides broadly and feebly arcuate, becoming straight toward base, the sur- face coarsely but rather sparsely punctate, becoming aeneo-punctate and rugose laterally and apically; scutellum scarcely one-half wider than long, oval, broadly impressed; elytra barely wider than the pro- thorax, parallel and straight at the sides to apical third, thence gradu- ally rounding to the emarginate apices, not at all serrulate except feebly very near the tip; surface not notably depressed posteriorly, sculptured as in planicauda though very much more sparsely and less conspicuously, the only regular series, except the sutural as in plani- cauda, being in four pairs; under surface black and aeneous, the abdo- men rugosely punctate and with large flattened punctureless patches laterally. Male with the apex of the fifth ventral transversely exca- vated, the posterior truncate edge broadly, angularly lobed toward the middle. Length 20.0 mm.; width 7.0 mm. Arizona (Nogales). cylindricus n. sp. 6 — Pronotum somewhat irregularly sculptured, generally with rounded punctureless areas anteriorly; body moderately narrow, cupreous, the elytra greenish-metalHc, parallel, feebly striato-punctate, the rounding sides posteriorly evidently serrulate, the surface without dense pubes- cence at base. Length 20.0 mm.; width 6.5 mm. Mexico. *mexicanus Lap.-Gory STUDIES IN THE AMERICAN BUPRESTIDiE 59 Prasinalia n. subgen. Hlod tarsus as long as the tibia. Body elongate, cuneate, a^neous-green, shining; epistoma truncate; head intricato-rugose, the antenna; very long, nearly one-half longer than the head and prothorax, the latter slightly wider than long, narrowed anteriorly, the sides nearly straight, truncate in front, "trisinuate" at base, the hind angles acute, not everted; surface coarsely, and, at the sides, confluently punctate; elytra narrow, convex, gradually narrowing from base to apex, the tips bidentate, the surface rather coarsely and irregularly striato-punc- tate; under surface finely and densely punctate, paler in color, the prosternum polished and violaceous; hind tarsi compressed, the first joint as long as the two following combined. Length 18.8 mm. California (Owen's Valley). [=Gyascutus cuneatus Horn]. cuneatus Horn Hind tarsus distinctly shorter than the tibia, though slender and subcom- pressed 2 2 — Body narrow, the elytra unevenly cuneate, rather convex, polished, bright metallic green above and beneath, the prosternum cuprascent, the posterior margins of the ventral segments steel-blue; head with the vertex elevated, the ocular slopes abruptly descending from longi- tudinal polished impunctate areas along the sides, the median parts more irregularly rugose, the epistoma very feebly and broadly sinuate; antennae extending to basal fourth of the prothorax, the third joint equal in length to the fourth, the eleventh with a short terminal process; prothorax trapezoidal, with the sides but very feebly arcuate, less than one-half wider than long, the apex arcuato-truncate, the base broadly sinuate laterally and with a small feeble sinus at the scutellum; surface moderately and sparsely punctate, more coarsely and coalescently so laterally, the basal angles acute; scutellum small, only slightly trans- verse; elytra not evidently wider than the prothorax, feebly narrowed to near apical third, then more strongly to the acutely ogival apex, the sides posteriorly serrulate, the apices emarginate; surface strongly and closely, and, toward the sides and apices, confusedly punctate, the punctured striae evident however and regular except near the sides, somewhat impressed; under surface finely but not very closely punc- tate, the prosternum medially polished and with a few punctures scat- tered along the median line. Length 19.0 mm.; width 6.0 mm. Ari- zona (Florence) angustus n. sp. Body larger and much stouter, similar in lustre and coloration, except that the ventral segments are not margined with blue, the entire abdominal surface irregularly clouded with metallic cupreous and green; head nearly similar, except that the elevated polished areas parallel to the eyes are not so well defined and are covered with coarse sparse punctures; antennae similar, the eleventh joint with a short terminal process; prothorax similar but more transverse, with the sides more arcuate, becoming more evidently sinuate basally, the basal margin at the scutellum not evidently sinuate and the surface broadly impressed 6o CASEY medially, broadly flattened toward base; elytra as wide as the prothorax, very feebly narrowed to near apical third, then much more rapidly and with the sides serrulate to the emarginate apices, the punctures everywhere well separated, the punctured series wholly unimpressed and almost regular though rather inconspicuous; under surface nearly as in angustus, the abdominal punctures fine, feebler and more unevenly distributed. Length 24.0 mm.; width 7.5 mm. Southern Califor- nia metallicus n. sp. If the same sexual difference in the antennae obtains in Prasinalia as in Hippomelas, the types of both angustus and metallicus are females; but I can scarcely imagine so great a sexual divergence in the antennae, and especially in the hind tarsi, as equality of either of them with cuneatus would imply, assuming the type of that species to be a male. The synonymy of mexicanus with saginatus, usually maintained, is I think wholly without basis of fact. There is a specimen before me, from Guerrero, that I have assigned to mexicanus, though it is more aeneous throughout above and lacks the rounded punctureless spaces on the pronotum. There is no material from the lower Rio Grande valley in my collection at present, so it is only possible to utilize the characters given in the orig- inal description for sphenicus Lee. It is without much doubt dif- ferent from any of the more western forms above described. Brevi- pes is almost subgenerically different from the remainder of Hip- pomelas and is placed first as being at one of the extremes in tarsal structure; the fifth ventral of its male is not transversely excavated at tip, though having the edge laminate; it is very broadly truncate at apex, with two small, widely separated and nearly circular emargi- nations. Gyascutus Lee. In this genus the body is convex, though less elongate and more oval than in Hippomelas, and the head is relatively much smaller; the sur- face sculpture is far rougher as a rule and the finely pubescent areas which, during life become thickly coated with a loose whitish fari- naceiform material, are more extensive, particularly at the sides of the under surface and at the base and along the depressions of the elytra; the finer sculpture and longer pubescence of these areas serve to retain the powdery coating more securely. The general structure is almost as in Hippomelas, though the prothorax is always strongly rounding at the sides and more narrowed anteriorly and the hind coxae are less STUDIES IN THE AMERICAN BUPRESTID^E 6l dilated internally, their posterior edge being less oblique. The sexual characters are much the same, the fifth ventral being more or less similarly truncate and transversely impressed in both sexes, sometimes with a minute tooth at each side of the truncature. The basal joint of the hind tarsi is frequently much longer than the next two combined, the epistoma broadly sinuato-truncate, the elytra invariably bispicu- lose at tip and the coloration of the body always metallic. The species are numerous, those known thus far being the following: — Pronotum finely and densely punctate between the smooth elevations even medially; elytra with finer and denser punctures in a longitudinal sublateral area; coloration generally greenish-metallic 2 Pronotum sparsely and generally more coarsely punctate medially and basally, the elytral sculpture more even, without a clearly demarcated sublateral area of much finer and denser punctuation; size smaller as a rule, the coloration generally coppery 11 2 — Elytra rapidly narrowed from base to apex, the latter relatively acute. Body rather stout, convex, shining, obscure greenish-metallic through- out, the depressions rather closely but briefly pubescent; head with large confused polished rugae, except broadly along the eyes and in front, the depressions densely punctate, senescent and densely pubescent ; prothorax about two-thirds wider than long, trapezoidal, the sides strongly arcuate, converging and straighter anteriorly, sinuate toward the prominent and acute basal angles, the base broadly sinuate laterally and very broadly and feebly at the scutellum; apex broadly, feebly arcuate medially; surface with numerous polished elevations, except in the apical transverse impression, all the depressions densely punc- tate; scutellum slightly transverse; elytra at the humeri much wider than the prothorax, the sides feebly arcuate; surface of each with three polished, somewhat ragged costae, the interval of the second and third broad and unusually finely, densely punctate; striation only traceable between the polished sutural elevation and the first costa; under sur- face unequally punctate, finely and densely toward the bases of the abdominal segments and on the hind coxa; prosternum narrowly and feebly impressed and punctate along the middle; hind tarei very slender though evidently shorter than the tibiae. Length 21.5-23.5 mm.; width 7.8-9.0 mm. California (Yuma). [=-Chalcophora planicosta Lee] planicosta Lee. Elytra not rapidly narrowed from base to apex, at the humeri less notably broader than the prothorax as a rule 3 3 — Eyes convex and notably prominent as in planicosta 4 Eyes feebly convex, not at all prominent, continuing the general outline curve of the head 10 4 — Elevated polished spaces of the pronotum very numerous and conspic- uous 5 62 CASEY Elevated pronotal spaces very small and few in number, widely scattered; sculpture and subsutural striation of the elytra aberrant 9 5 — Elytra distinctly subcostate, cuprascent. Body moderately stout and convex, shining, the pubescence of the upper surface and head unusually short, also fine, short and close beneath; head and pronotum bright aeneous, the former closely and subevenly rugose, except near the eyes and on the occiput, the inter-rugal punctures coarse and not dense; antennae ( d^) extending rather beyond the base of the prothorax, the latter three-fifths wider than long, very slightly narrower than the elytra, of the usual form, the basal angles distinctly everted and acute; scutellar sinuation feeble; surface with strong but very irregular pol- ished elevations, continuous along the median line, the interstitial punc- tures rather small and dense; elytra just visibly narrowed and with nearly straight sides to apical two-fifths, thence obliquely to the rather acute apex; surface with the three broad costae of planicosta very evi- dent but still more ragged and broken, the striae, except the sutural and scutellar, scarcely traceable; punctures close, finer and denser laterally. Length 20.0 mm.; width 7.7 mm. Arizona (Bill William's Fork), — F. H. Snow snowl n. sp. Elytra with the costae never more than traceable; coloration generally more greenish 6 6 — Form narrower and subparallel, the sinus of the elytral apices broader than usual 7 Form very broadly suboval 8 7 — Lustre moderately shining, greenish-aeneous; pubescence of the elytra short but rather abundant, subvittate; head subcupreous, unusually densely pubescent, rather finely, very densely punctate throughout, the median line carinate medially, the elevated spots very few in num- ber and inconspicuous; antennae (c?) extending slightly beyond the base of the prothorax, unusually dilated from the base of the fourth joint; prothorax three-fourths wider than long, the sides broadly round- ing, evidently more so medially, the basal angles prominent; surface with strong and confused through isolated elevated spaces, the punc- tures of the interspaces strong, dense and cribrate; elytra between three and four times as long as the prothorax and a little wider, the dorsal humeral swelling pronounced; surface striated, suturally, the costae evident but only feebly convex, not elevated, the surface laterally deeply and very densely cribrate; under surface greenish-aeneous, finely, densely punctate except medially. Length 23.5 mm.; width 8.8 mm. Arizona cribriceps n. sp. Lustre strongly shining, greenish, sometimes with a faint xneous tinge; pubescence less abundant, very short; head metallic-green, confusedly rugose, the punctuation of the interspaces coarse, close but much less extensive than the smooth rugosities, the pubescence much sparser than in cribriceps; antennae not so broad at the fourth joint; prothorax nearly similar but with the small punctures of the interspaces not so dense or deeply cribrate; elytra but little wider than the prothorax, STUDIES IN THE AMERICAN BUPRESTID^ 63 the sides straight and parallel to apical two-fifths, thence rapidly rounded and oblique to the rather obtuse apex, the humeral swelling not conspicuous; disk without costa?, their positions indicated by greater convexity and sparser punctures of the surface, which is evidently striate suturally, finely and only moderately closely punctate even sublaterally, the two apical spines unusually developed. Length 15.7- 21.0 mm.; width 5.7-7.8 mm. New Mexico (Demming). pistorius n. sp. Lustre moderately shining, obscure greenish-metallic, the surface more convex than in either of the preceding species and with the head decid- edly smaller, the large polished rugosities more extensive than the punctuation; antennae {d") with the third joint barely as long as the fourth, the latter and succeeding joints moderately compressed, parallel except basally; prothorax two-thirds wider than long, of the usual form, much narrowed anteriorly, the basal angles abruptly everted, acute and very prominent; surface covered with strong polished confused elevations, the punctures of the interspaces rather coarse, fine along the basal margin; elytra subparallel, the sides nearly straight for three- fifths, thence rounding, oblique, and becoming nearly straight to the apex; surface nearly even and almost equally convex throughout, the sublateral punctures not very close-set; striae obliterated, except the sutural and three more visible double striae on each, also a fourth visible apically near the sides and more tumid; under surface irregularly and, for the most part, finely and closely punctate. Length 20.0 mm.; width 7.3 mm. Arizona (Winslow),— H. F. Wickham. acutangulus n. sp. 8— Surface metallic-green with feeble aeneous tinge, only moderately shining; pubescence ashy as usual, rather distinct, feebly subvittate on the elytra; head nearly as in pistorius but with fewer rugaj and greater extent of dense punctures, particularly before the middle, closely and conspicu- ously pubescent; antennae (c^) nearly as in pistorius but with the third joint much more elongate, being decidedly longer than the fourth; prothorax short, nearly four-fifths wider than long, rounded at the sides, unusually narrowed but with arcuate sides to the apex, the basal angles unusually feebly everted; surface nearly as in pistorius; elytra evidently wider than the prothorax, the sides very feebly converging and feebly arcuate for three-fifths, thence slightly more arcuate and gradually strongly convergent but arcuate to the tip; surface nearly as in pistorius; under surface more unevenly punctate. Length 22.5 mm.; width 8.7 mm. Arizona pimaUcus n. sp. Surface much more polished and pale brassy-green in color, generally not at all cupreous, feebly pubescent, the pubescence inconspicuous and scarcely at all vittate on the elytra; head in great part coarsely and closely rugose, rather sparsely pubescent; antenna; (c?) with the third joint slender but scarcely as long as the fourth; prothorax very trans- verse ( 9) or much less so (c?), the sides more arcuate but rather less converging anteriorly than in pimalicus, the hind angles much more prominently everted and acute; surface nearly similar, except that the Proc. Wash. Acad. Sci., April, 1909. 64 CASEY punctures are finer, less dense and more inconspicuous medially toward base; elytra almost evenly convex, without trace of costae except feebly toward tip, the surface evidently though finely striate and more coarsely sculptured in about inner half, the fine punctures sublaterally not close- set and rather uneven; outline nearly as in pitnalicus; abdominal seg- ments finely and densely punctate only toward base. Length 19.3- 24.5 mm.; width 7.0-10.2 mm. Utah (St. George), — H. F. Wickham. amplus n. sp. A — Body more elongate, bright brassy-cupreous in color; head nearly similarly rugose but more finely, closely punctate toward the eyes and on the occiput; prothorax twice as wide as long, nearly as in amplus but wider at base, the basal angles not so acute or everted but extending sensibly beyond the elytral humeri; elytra more elon- gate, with the sides very rounding and converging posteriorly, very much more gradually so than in amplus, the surface less rugose. Length 23.5 mm.; width 9.0 mm. Utah (St. George), — H. F. Wickham aerata n. subsp. 9 — Very stout, elongate-oval in form, convex, only moderately shining, bronzy-aeneous in color, the pubescence of the elytra very short, sparse, evenly distributed, without trace of vittate appearance; head not densely pubescent, about equally rugose and densely punctate; antennae (9) attaining about basal third of the pronotum, the prothorax large, but little narrower than the elytra, two-thirds wider than long, the sides subevenly rounded, becoming strongly convergent anteriorly, the basal angles everted and acute; surface with the median line, which is irregu- larly enlarged anteriorly, and a few small scattered spots elevated and smooth, the major part of the surface, however, rather coarsely, deeply and densely cribrate; elytra with the sides just visibly converging and slightly arcuate for three-fifths, then rather rapidly rounding, becoming oblique and feebly arcuate to the apex; surface subequally and densely punctate throughout, the striae especially evident in four pairs on each, the first pair not evenly arcuate to the tip but becoming more strongly and externally arcuate in about apical fourth; under surface in great part finely though not densely punctate, very briefly and inconspicu- ously pubescent. Length 24.5 mm.; width 9.8mm. Utah (St. George), — H. F. Wickham egregius n. sp. 10 — Elytra with obsolescent smooth costae or very evident vestiges of dis- integrated costulation, the surfaces between the very feeble elevations somewhat concave and finely, densely punctate except suturally; body moderately stout, parallel, rather convex, greenish or viridi- aeneous to feebly cuprascent and rather shining, the pubescence dis- tinct and somewhat longer than usual; head rather smaller than usual, the eyes brown in all examples examined, the front about equally rugose and finely, densely punctate; antenna moderate, the third joint barely as long as the fourth ( c?) or a little longer( $ ) ; prothorax two- thirds wider than long, narrowed anteriorly, arcuate at the sides, the basal angles very variable in prominence and eversion; surface with finer, dense punctures and large irregular polished elevations, the me- STUDIES IN THE AMERICAN BUPRESTIDiE 65 dian line broadly and irregularly convex and smooth, more or less evi- dently bifurcating anteriorly; elytra but little wider than the prothorax, the sides just visibly converging for three-fifths, then rounding and arcuately converging to the tip, the striation evident suturally; under surface finely, densely punctate except medially, the abdominal seg- ments coarsely and sparsely so apically as usual. Length 16.0-23.0 mm.; width 5.5-8.5 mm. Western Texas (abundant near El Paso). [ = Chalcophora ohliterata Lee] obliteratus Lee. Elytra without trace of the principal costules of obliteratus, evenly convex and more broadly, subevenly striate, with coarsely broken interspaces in inner half of each, the surface thence to near the sides concave and closely punctate though much more coarsely and less densely than in obliteratus; size smaller, the form more slender, shining, brassy-green to dark green in color; head generally with three lines of coarse polished broken ruga?, the remainder densely and finely punctate; antennae moderate; prothorax nearly similar but shorter, the apical margin generally more angulate at the middle, the basal angles acute and slightly everted, the scutellar sinus broad and very feeble; elytra nearly as in obliteratus in general form, the under surface unequally punc- tured as usual. Length 14.7-17.0 mm.; width 5.0-6.1 mm. South- eastern Arizona fldelis n. sp . II — Eyes with their inner margins strongly converging upward as usual; basal angles of the prothorax prominent and acutely everted 12 Eyes with their inner margins but feebly converging upward, the basal thoracic angles not everted 17 12 — Prothorax very evidently narrower at apex than at base 13 Prothorax about as wide at apex as at base; body very small in size 16 13 — Body fusoid, the head not or but little wider than the base of an elytron in either sex 14 Body subcylindric, the head notably wider than the base of an elytron 15 14 — Form very stout, convex, rather shining, deep cupreous in color, the ashy pubescence short but distinct, evidently vittate on the elytra; head unevenly punctate, finely along the eyes, for the most part, how- ever, coarsely and confusedly rugose; eyes large and slightly prominent; antennae short ( $ ) , with the third joint much longer than the fourth, the remaining joints strongly serrate; prothorax two-thirds wider than long, strongly narrowed anteriorly, rounded at the sides, the basal angles only slightly everted; surface with strong and irregular eleva- tions, with densely punctate interstices laterally, coarsely, rather sparsely and unevenly punctate medially and basally, the scutellar sinus very broad and feeble; elytra evidently less than twice as long as wide, but little wider than the prothorax, of the usual subparallel form, obliquely narrowed posteriorly, the surface throughout subequally convex and rugose and unevenly, not densely punctate, with the striation more or less evident though deeper suturally; fine punctures of the under surface less dense than in the preceding section, the pu- 66 CASEY bescence very short and inconspicuous. Length ( 9 ) 20.0 mm. ; width 7.8 mm. Texas (near El Paso) obesus n. sp, A — Body nearly similar in general form but much narrower and with less widely separated eyes; — allowing for differences in sex prevailing elsewhere, — and dififering also most obviously in being wholly aeneous- green, without trace of the strong cupreous color of obesus; elytra similarly narrowly and deeply sinuate at the tips and strongly bispic- ulose, the very short pubescence even more evidently arranged in broad and moderately definite vittae. Length (d') 17.5 mm.; width 6.5 mm. Texas (El Paso) aeneoviridis n. subsp. Form rather stout though evidently less so than in the preceding, cupreo- aeneous to deep coppery-red, shining, the pubescence very short and sparse, not vittate on the elytra; head rather concave anteriorly, densely punctate, more or less rugose posteriorly on the vertex; eyes much less prominent, their outline continuing that of the head, becoming more longitudinal posteriorly, pale brown with fine black margin in all the numerous examples at hand; antennae nearly as in obesus but more slender; prothorax nearly as in obesus but less narrowed anteriorly and with the basal angles more strongly everted and acute, the scutellar sinus a Uttle deeper and more angular; elytra nearly similar through- out though a little more elongate, almost twice as long as wide ( J') or somewhat shorter ( 9 ) ; under surface nearly similar and with numerous coarse perforate punctures along the middle of the proster- num and densely throughout the width anteriorly. Length (cJ*) 12. 5-16.5, (9) 13. 3-17. 5 mm.; width (c?) 4.0-6.2, (9) 4.6-6.8 mm. Southeastern Arizona compactus n. sp. A — Body nearly as in compactus but larger and more oblong, bright cupreous, the head slightly larger and with the eyes still more widely separated; prothorax nearly similarly sculptured but more trans- verse; elytra more evidently oblique at the sides in apical third, the pubescence less uniformly distributed, being rather obviously in single lines but without the broader vittae of obesus and ccneoviridis. Length (9) 19.0 mm.; width 7.1 mm. Southeastern Arizona. solidus n. subsp. Form moderately robust, reddish or coppery- bronze, moderately shining; prothorax with the sides moderately arcuate, the disk convex, coarsely punctured, somewhat tuberculate at the sides from confluence of the punctures but without distinct callosities; scutellum small, oval; elytra densely and deeply punctate, the punctures near the apex less confused and having a slight tendency to form striie; body beneath more shining than above, moderately densely punctured but smoother along the median line; presternum very coarsely punctured in front, the j^rocess smooth, with a row of coarse deep punctures so closely placed as to resemble a groove. Length 15. 0-19.0 mm. North Carolina (near Wilmington), — fide G. H. Horn carolinensis Horn 15 — Body rather narrowly, subcylindrically convex, shining, cupreous, the pubescence short but very evident, not vittate on the elytra; head for STUDIES IN THE AMERICAN BUPRESTID^ 67 the most part coarsely, closely and confusedly rugose, finely punctate along the eyes, only slightly concave anteriorly, the eyes neady as in compacius but slightly more prominent, the antennae nearly similar (d^); prothorax much narrower, not quite one-half wider than long, otherwise similar, only moderately narrowed anteriorly, with the apex but very feebly arcuate medially, the basal angles only feebly sub- everted, the scutellar sinus feeble; surface similar but more coarsely as well as more sparsely punctured medially; elytra not quite twice as long as wide, evidently wider than the prothorax, the posterior ob- liquity slighter and more gradually formed, the surface sculpture almost similar but becoming more rugose laterally; under surface polished, brighter coppery- red, the fine punctures nowhere dense; presternum with the very coarse perforations widely and irregularly spaced along the middle. Length 15.0 mm.; width 5.6 mm. New Mexico. cylindrinus n. sp. 16 — Form very narrowly suboval, moderately convex, shining, bright bronze above, feebly cuprascent beneath, the pubescence very short, sparse and inconspicuous, not vittate on the elytra; head punctate and also with flattened and confused smooth spaces, only feebly concave ante- riorly; eyes and antennae nearly as in compacius; prothorax scarcely more than one-half wider than long, the sides subparallel, feebly arcu- ate medially, rather abruptly and subsinuately convergent near the apex, the basal angles not at all everted; surface coarsely, confluently punctate laterally, finely and sparsely so and basally depressed medially and with two divaricately oblique impressions from the middle of the base; scutellum very small; elytra uniformly convex and sparsely punctate and rugulose throughout, the striae fine, distinct but not im- pressed suturally, a submarginal costule evident for a short distance near the middle; under surface not finely, rather sparsely and subrugu- losely punctate laterally; anterior tibiae (J*) feebly arcuate, scarcely serrate, the middle tibiae almost straight. Length 9.2 mm.; width 3.0 mm. Southeastern Arizona debilis n. sp. Form|more elongate and subcuneiform, shining, deep coppery- red ; head and antennae ( (^) nearly as in debilis; prothorax rather less than one-half wider than long, the sides broadly rounded, very gradually converging anteriorly to the apex, which is very slightly narrower than the base, feebly prominent medially, the basal angles very slightly everted, acute, the scutellar sinus unusually deep and distinct; surface broadly de- pressed basally but not bi-impressed, coarsely, confluently punctate laterally, very sparsely medially; elytra more elongate, much more than twice as long as wide, evenly convex, finely, very sparsely punc- tured and sparsely rugulose throughout, the striation barely evident even suturally; under surface not densely, subrugulosely punctured laterally; anterior tibiae (c?) distinctly arcuate and remotely serrulate within, the middle tibiae also arcuate though more feebly, but serrulate within, only near the tip; posterior tibiae nearly straight. Length ii.o mm.; width 3.5 mm. Southeastern Arizona, — G. W. Dunn. tenuis n. sp. 68 CASEY 17 — Form and coloration nearly as in fom/)or/«A- but much smaller, metallic- purplish, shining, the pubescence fine, short and sparse; head coarsely, unevenly and coniluently punctate; antennic (?) unusually short; prothorax moderately transverse, convex, narrowed anteriorly, irregu- larly and coarsely j)unctate, the punctures conlluent laterally with large rugosities, the broad median s])acc simply irregularly and rather sparsely pumtate; hind angles right; elytra of the usual form, having small irregular smooth places and also finely ])unctate, the punctures somewhat regularly serial suturally but confused laterally and apically, where they become somewhat transversely continent though still not dense; meso- and metasternal side-j)ieces not coarsely but subcon- fluently punctate; abdomen rather finely, subrugosely punctured at the sides; anterior tibia' very slightly arcuate, the intermediate and j)os- terior nearly straight. Length 11. 5-13.0 mm. Utah (Thadburn's Ranch) juniperinus Wick. The sexual characters in this genus are feeble, and, although the male is generally a little smaller and narrower than the female — sometimes notably so, the readiest way to distinguish the sexes is by the structure of the antenna', all the joints being elongate in the male, the outer ones becoming raj)i(lly shorter in the female. The anterior, and sometimes also the intermediate, tibiie are arcuate, being more or less serrulate within in the male, but the anterior are simply more or less bent in the female. My cotype of juuipcrhius, very kindly sent to me by Prof. Wickham, is a female measuring 11.5 by 4.0 mm. in size; the species is peculiar in having the internal margins of the eyes so feebly converging posteriorly, this character having been first noticed by the original de.scribcr. The locality of carolincnsis Horn, is open to the gravest doubt, esjiccially when we consider its perfectly normal type of form, coloration and sculpture. I am of the opinion that no species of Gyascutus occurs beyond the desert Sonoran prov- inces. On comparing compactus with what purported to be a speci- men of carolincnsis in the LeConte cabinet, I found the latter to be very similar, though rather less robust and with tiio scant i)ubcscence more vittate in arrangement on the elytra. In comj>arison with Ilip- pontclas, Stictoccra and Spinthoptcra, where individual stature appears to be rather constant, the body in Gyascutus frequently varies enor- mously in size within specific limits; indivieluals are usually abundant when they occur at all. In the first section of the genus the broad and densely punctate imjiression ali)ng the aj>ical thoracic margin is much more developed and abruptly defined than in the second section. STUDIES IN THE AMERICAN BUPRESTID.E 69 Stictocera n. gen. While similar to Gyascutus in general organization, the compara- tively few species of this genus (HfTcr in several important structural features. The hind coxx% for example, are more rapidly though grad- ually expanded internally as in Hippomelas, and the tarsi also are nearly as in the latter genus. In the pronotum Stictocera differs very much from either Gyascutus or Hippomelas, having the surface even in general convexity, as in the latter, but with a different type of sculp- ture and without the smooth irregular callus along the median line so very general in the former, so that the elevated basal margin, which distinguishes the genus from all its associates, is entire and not discontinuous at the middle. The species are frequently of large size, comprising the largest Buprestids of this country, and sometimes of very broad elongate-oval convex form, though also occurring nar- rower and more parallel. The antennae are strongly compressed, sub- parallel and densely punctate distally, the outer joints with the apical margin finer than in Hippomelas and with the edge sharply de- fined, the surface internally obliquely merging into the general sur- face of the joint, the sensory fossa sharply defined, elongate-oval or feebly reniform and near the articulatory foramen. The elytra are evenly convex, uniformly though rugosely sculptured and obscurely striate, as in the second group of Gyascutus, the sides seldom very evidently serrulate posteriorly and the apex distinctly bispinulose. The species may be distinguished as follows : — Basal joint of the hind tarsi but little longer than the second. Head brassy- punctate, with elevated reticulate lines; prothorax with deep punctate aeneous ruga^, less densely sculptured than in Gyascutus obliteratus, the polished elevations cyaneous, narrowed anteriorly, the sides broadly rounded, the surface transversely impressed at apex, the base Insinuate, the posterior angles acute; elytra striato-punctate, transversely and sparsely but dee[)ly rugose, the sides subserrate posteriorly; surface "almost constantly with a large brassy patch on each elytron. " — [fide Waterhousej; under surface aeneous, densely punctate and pubescent; pectus and posterior segments of the abdomen coarsely punctate, shining, variegated with bluish reflections. Length 23.3 mm. Mexico (Ures, in the State of Sonora) . [ = Chalcophora ccelata Lcc] . . caelata Lee. Basal joint of the hind tarsi much longer than the second, though distinctly shorter than the next two combined 2 2 — Form broadly suboval when well developed, varying greatly in size, shining, dark steel-blue, the depressions brassy; pubescence short. 70 CASEY sparse, evenly distributed on the elytra, coarse and erect, longer and dense in isolated patches on the under surface; head more densely pubescent, with large irregular smooth rugae, finely, densely punctate along the eyes and in an angulate anterior area; eyes large, prominent, brown in all examples at hand; antennae not quite attaining the base of the prothorax ( $) or longer (c?), the third joint subequal in length to the fourth; prothorax about twice as wide as long, four- fifths wider than the head, strongly narrowed anteriorly, the sides arcuate, the basal angles everted and acute, the scutellar sinus broad and deep; surface somewhat flattened medially toward base, vermicularly eroded and punctate, the low flat smooth areas large, becoming strongly ele- vated laterally; elytra distinctly wider than the prothorax, barely twice as long as wide, the sides moderately converging for three-fifths, thence obHque and but slightly arcuate to the apex; surface subuniform throughout, the striae moderately distinct almost to the sides, the inter- vals much interrupted by subquadrate spots of greater convexity; abdominal segments steel-blue, very coarsely, unevenly and remotely viridi-punctate, finely, densely punctate only near their bases laterally. Length 21. 0-31.0 mm.; width 7.2-12.6 mm. Texas (near El Paso). pollens n. sp Form narrower and subparallel, similarly convex and with the same col- oration, lustre and sculpture; head and eyes nearly similar, except that the latter are rather more strongly convex and decidedly less widely sep- arated; antennae parallel and strongly compressed, the lower part of joints four to eleven pale in color as usual and with the lower margin rectilinearly truncate except basally, differing very markedly in the sexes, as in Gyascutus, extending well beyond the base of the prothorax ( cJ*) or failing to attain the base by a corresponding amount ( 9 ) ; pro- thorax not so transverse as in pollens but otherwise similar; elytra narrower and more parallel, about twice as long as wide, slightly wider than the prothorax, the sides arcuately converging in apical two-fifths, with sculpture and sparse erect seta^ as in pollens, the under surface nearly similar; anterior tibiie (c?) slightly bent and with a few very minute serrules within apically, the intermediate more strongly ser- rulate apically but almost straight. Length 22.0-26.0 mm.; width 8.0-9.7 ^^- Arizona (Tufson), — H. F. Wickham. . . laticornis n. sp. The single very small male of pollens in my series has the anterior tibiae evenly bowed but with the inner subapical serrules almost com- pletely obsolete, the intermediate distinctly arcuate, though less strongly, but with the inner subspiniform serrules from behind the middle to the apex much stronger. Laticornis differs from pollens, not only in its narrower form and smaller size on the average, but in its more convex and rather less widely separated eyes, with their inner margins somewhat more strongly converging upward and also in its evidently less transverse prothorax. The apex of the last STUDIES IN THE AMERICAN BUPRESTID^ 7I abdominal segment is thick, truncate and transversely excavated throughout the width of the truncature in both sexes. Spinthoptera n. gen. The genus Psiloptera, of Solier, having as its type the South Ameri- can attenuata, of Fabricius, as amplified by Lacordaire became very composite and indefinite in its scope, a number of generic types being included by him which have since been separated and held to be amply valid, such as Perotis, Polybothris, Hippomelas and Ectinogonia. Thus simplified Psiloptera has been continued by various authors without much further amputation of members, and I have ventured here to detach those forms having the prosternum truncate anteriorly and not medially prolonged and the elytra studded with shallow, finely punctate and pubescent metallic areolae, under the generic name Spinthoptera. General facies, completely ignored by Lacordaire, is of itself in reality the most important structural character prevailing in this group of genera and is more decisive than modifications of special organs, of which there are very few. In fact the genera might be termed opinionative, in the same sense that species not character- ized by striking structural differences in special organs are usually called opinionative species. To maintain that either species or genera are less real for being of such an opinionative nature is not justifiable; it merely necessitates greater intelligence and experience in specifically delimiting such series of individuals, or in aggregating such groups of species to form genera, with a greater attendant risk of mistakes; but after all, in the search for truth, which alone should be the ulti- mate aim, errors of judgment are practically unavoidable at times. We note in this genus a departure in several structural features from those of the three preceding genera, aside from the special and peculiar modification of the antennae common to all the Psilopterids. There is, for example, no terminal antennal process; the prosternum, instead of being punctured or feebly impressed along the middle, is remotely bisulculate, and the first abdominal segment, instead of being uniformly convex medially, is here either flattened or sulcate. We should note also an entire absence of the raised basal margin of the pronotum and ante-scutellar sinuation of the Hippomelids, the con- stant presence of two small perforate ante-scutellar discal punctures, wholly unsuggested in the Hippomelids but forcefully remindful of 72 CASEY Dicerca, and the frequent posterior prolongation of the elytra in Psiloptera, also betraying an obscure bond of affinity with that genus and presumably signifying derivation from a common primitive stock. The striation of the elytra is much more developed and obvious than in the Hippomelids, the hind coxas rapidly dilated within as in Hip- pomelas and Stictocera, the outer antennal joints much compressed, rectilinearly truncate beneath as in those genera and wholly differing from Dicerca, densely punctulate along their obliquely flattened lower edges and with sensory fossa; lateral and linear in form, as well as one terminal, which is narrow, deep and oblique. The hind tarsi are thick, with the three basal joints progressively diminishing very slightly in length; sexual differences are virtually unobservable. The species and subspecies of our fauna may be outlined as follows: — Prothorax widest at or near the middle; elytra arcuately narrowing pos- teriorly and more or less obtuse 2 Prothorax widest at base 6 2 — Prothorax rounded at the sides; first ventral segment merely flattened along the median line 3 Prothorax angulate at the sides at or slightly behind the middle; first ven- tral segment deeply sulcate along the median line 5 3 — Antennae metallic blue or greenish throughout. Body large in size, stout, convex, shining, dark steel-blue, the impressions bright greenish- aeneous, subglabrous, the elytral areolae densely punctate and having short decumbent pubescence, the irregular sparse vestiture of the under surface longer, decumbent, arranged in sparse radiating clusters of three or four hairs each from the very coarse transverse punctures, especially of the abdomen; head very coarsely rugose throughout, the antennas short, but little longer than the head in either sex, slightly more apically attenuated in the male; eyes large though only moder- ately prominent, widely separated; prothorax barely twice as wide as long, widest behind the middle, the sides broadly, subevenly rounded, becoming gradually more convergent apically and feebly so and slightly sinuate basally, the basal angles right; apex feebly arcuate except later- ally, the base very broadly lobed, sometimes subangularly; surface convex anteriorly and laterally, more flattened basally and with a dis- tinct subquadrate impression at the middle near the base; punc- tures very coarse and irregular, generally sparce medially, verj' unevenly and vcrmicularly contluent laterally; scutcllum small, rounded; elytra but Httle wider than the prothorax, not quite twice as long as wide, Ijaraflel, gradually rounding at the sides and obliquely, subarcuately narrowed in apical two-fifths to the apices, the latter narrowly truncate, sometimes obliquely with the sutural angle more prominent; punctures large, impressed, arranged in regular series which are slightly impressed. STUDIES IN THE AMERICAN BUPRESTID.E 73 not much more so laterally, the finely punctate shallow areolae rather small in size, frequently subtransversely coalescent; prosternum densely, coarsely punctate, broadly polished and impunctate along the middle; legs blue, the femora frequently subviolaceous. Length 26.0-28.8 mm. ; width 10. 2-1 1.6 mm. Texas (El Paso) convexa n sp. Antennae probably metallic blue in color but not described, the body nearly as in convexa but with the prothorax very much more transverse, the form stout, the color dark steel-blue, the head and prothorax with greenish-brassy punctures, the latter unequally punctate, at the middle basally broadly foveate, almost three times as wide as long, broader before the base, the sides anteriorly strongly rounded, posteriorly sinuate, the hind angles right; elytra with punctured striae, the inter- vals with many quadrate spaces, which are impressed, greenish-brassy, granulate and pubescent, at the apex obliquely subtruncate, the sutural angles rather prominent; under surface densely and coarsely viridi- punctate; prosternum bisulcate. Length 22.5-28.7 mm. Texas (Eagle Pass,— LeConte). [ = Psiloptera valens Lee] valcns Lee. Antennae non-metallic, except sometimes feebly toward base 4 4 — Body stout, parallel and convex, nearly as in convexa and valens in color- ation,^ lustre and sculpture throughout, the eyes larger, much more prominent and less widely separated than in the former; antennas simi- lar but deep black, becoming faintly bluish toward base; prothorax shorter than in convexa, apparently somewhat more than twice as wide as long and of a different outline, being more anteriorly inflated and widest before the middle laterally, more sinuate basally, the angles similar, the basal impression more extended anteriorly, obsolescent beyond the middle; elytra nearly similar but scarcely more than three- fourths longer than wide, the apical truncature oblique, angularly acute and prominent suturally, the aeneous areolae larger, the punctures of the series toward the suture smaller; under surface nearly similar, the tibia? rather more closely punctate and pubescent, the abdomen with large and uniform dense patches of pubescence laterally. Length 23.5-24.5 mm.; width 9.2-9.8 mm. Western Texas (locality unre- corded) ocularis n. sp. Body more slender, parallel, convex, very much smaller in size, shining, almost similarly sculptured; head small, black, with coarse aeneous rug^, the eyes moderate, not notably prominent ; antennae short, piceous- black throughout; prothorax black, convex, three-fourths wider than long, only moderately and gradually narrowed anteriorly, widest near basal third, the sides very broadly rounded, becoming sinuate only at the basal angles, the surface deeply and rather abruptly impressed at the middle basally, the punctures very coarse, not close, aenescent and irregular; elytra blackish-blue, in form and sculpture nearly as in the preceding species, the apices more narrowly and obliquely truncate; under surface nearly similar, except that the more densely and evenly pubescent sublateral patches of the abdomen are much more finely and feebly sculptured; legs and tarsi more slender, blue, the femora 74 CASEY very coarsely punctate, partially greenish basally. Length 15. 7-16. 2 mm. ; width 5.4-5.9 mm. Texas (locality unrecorded) parva n. sp. Body nearly as in parva but larger and stouter and differing from any of the preceding in the shining blue and coppery iridescence of the elytra, parallel, convex; head rather small, the eyes convex and prom- inent, brown in all specimens at hand, widely separated, the front with strong confused cupreo-asneous rugae; antennae shining, sometimes with faint greenish-black reflection, bluish toward base; prothorax not quite twice as wide as long, widest well behind the middle, nearly as in convexa, the median basal impression stronger and extending further anteriorly, generally evanescent beyond the middle, the smooth areas blue-black, the remainder aeneous; elytra nearly as in the pre- ceding species but with much larger and more coalescent cupreous areolae, which are more coarsely and densely punctate, slightly opaque and with short, less conspicuous pubescence, deeper laterally, where they produce a more rugose appearance; apices narrowly and ver)' obhquely sinuato-truncate, the truncature sometimes difficult to trace; under surface with the densely pubescent patches of the abdomen much more finely and densely sculptured than in convexa or ocularis but more nearly as in parva; legs bluish or bronzed, the femora, and the tibiae less evidently, deeper blue apically. Length 19. 4-2 1.5 mm.; width 7.0-8.0 mm. Northern Texas and Kansas. [=Dicerca wood- housei Lee] vvoodhousei Lee. 5 — Form oblong, stout, moderately convex, the pronotum much less strongly so anteriorly and laterally than in the preceding group and more grad- ually and strongly flattened throughout the width basally, but not or scarcely impressed at the middle subbasally, the two ante-scutellar punctures thereby rendered very obvious, blue-black, shining, the depres- sions throughout brilliant viridi-asneous; head strongly, vermicularly rugose and coarsely punctured, the eyes large and strongly convex, widely separated, the antennae and legs as in convexa, bright violaceous-blue; prothorax somewhat more than twice as wide as long, broadly angulate at the sides, widest distinctly behind the middle, where the width is evidently greater than at base, the converging sides nearly straight from angulation to apex, feebly sinuate basally, the basal angles acute ; surface with coarse aeneous punctures, sparse and unevenly distributed, becoming densely and very irregularly aggregated laterally; elytra nearly as in convexa but not so convex, the feeble obUque apical truncature variable, some- times transverse; punctures strong, arranged in coarse impressed series deeper laterally, the depressed areas brilliantly shining, densely, rather finely punctate and moderately pubescent, very variable in extent, generally isolated, sometimes larger and in great part subtransversely confluent; sulcus of the first ventral abrupt and deep, evanescent near the hind margin though in very variable degree, generally narrower, deeper and longer (cJ*) or rapidly shallow with more posteriorly flaring sides (9). Length 22.0-31.0 mm.; width 8.0-1 1.7 mm. Arizona. arizoniea n. sp. Form nearly as in "Ducrca divaricala^' but without elytral prolongation STUDIES IN THE AMERICAN BUPRESTIDiE 75 and more convex, black, shining above, the under surface bright cu- preous; depressions of the upper surface cupreous; head cupreous, the front feebly depressed medially, coarsely sculptured; antenna; short, obscure cupreous; prothorax convex, twice as wide as long, apparently widest slightly behind the middle, the sides subangulate, converging anteriorly, sinuate basally, the basal angles right; surface coarsely, sparsely and irregularly punctate, the punctures more crowded laterally, longitudinally and feebly impressed along the middle, with a narrow smoother median line; elytra gradually narrowing to the apices, the latter obliquely truncate, with the outer angle dentiform; serial punc- tures becoming transversely confluent laterad, producing a rugose ap- pearance, the intervals with a few punctures and some impressed punc- tured bright cupreous spots; under surface coarsely and more or less confluently punctate, the prostemum as usual; first ventral sulcate medially, the sulcus abruptly defined; femora cupreous, the tibiae black above vnth coarse greenish punctures, greenish-cupreous on the sides, the tarsi obscurely cupreous. Length 16.5-22.0 mm. Texas (Brownsville) . [ = Psiloptera cupreo punctata S.] . cupreopunctata Schaef . 6 — Form stout, steel-blue, the head and pronotum variegated with viridi- aeneous punctures, the pronotum unequally punctate, broadly impressed "foveato" in the middle at base; prothorax almost twice as wide as long, narrowed anteriorly, the sides at the middle obtusely angulate, thence feebly subsinuate to the base, the basal angles right; elytra with punc- tured striae, the intervals with quadrate impressed spaces, which are granulate, pubescent and greenish-golden in color, very variable in extent as in arizonica, obliquely feebly truncate at apex; under surface densely and coarsely viridi-punctate, the prostemum bisulcate. Length 26.5-31.3 mm. Mexico (Ures, in the State of Sonora). [=Psiloptera wehbi Lee] webbi Lee. Form short, stout and very convex, only moderately shining, blackish- bronze above with metallic green depressions; under surface dull viridi-aeneous laterally, deep and brilliant metallic roseate along the middle, very bright on the polished sterna, dull on the abdomen; head very coarsely, rugosely punctate, the antennal ridges unusually strong; eyes prominent; antennae black, very feebly bronzed, with the lateral sensory fossae small and rounded or oval, not Hnear as in the preceding species; prothorax glabrous, about twice as wide as long, the sides rapidly converging from base to apex, feebly angulate behind basal third, thence straight to the apex, slightly sinuate in less than basal third, the angles slightly less than right; surface coarsely, sparsely but irreg- ularly aeneo-punctate, more densely laterally, where there are some elevated smooth areas, very feebly and indefinitely impressed toward the middle of the base, the two punctures very evident; elytra but little wider than the prothorax, the sides parallel and straight for nearly three-fifths, thence gradually oblique, becoming very feebly sinuate near the apices, which are rather broadly and obliquely truncate, with the outer spine rather strong; surface with suturally feebly, laterally strongly, impressed series of coarse punctures, becoming contiguous 76 CASEY and foveiform laterally, also with deeply impressed, finely, densely punctate and puberulent green areolae, which become obsolete in about inner half, except posteriorly, where they extend to the suture through small in size; intervals punctureless; under surface closely, coarsely and irregularly sculptured, the sterna broadly smooth along the middle, the first ventral narrowly and deeply sulcate; legs bronzed, the femora with a feebly rose-colored reflection apically. Length 22.0 mm.; width 8.5 mm. Panama (Taboga Island) *insularis n. sp. Some of the above species, notably those of the valens series, appear to resemble the Mexican drummondi Lap. -Gory, in general form and sculpture, but the figure of that species seems to indicate a broader and more arcuate basal pronotal impression, and, as our forms consti- tute several allied species, I assume that none of them is actually synonymous with drummondi. It is impossible to imagine any reason whatever for assigning such mutually distinct species as ivehhi, valens and woodhousei to drummondi as varieties, and it is still more prepos- terous to consider them synonyms. The structure of the base of the abdomen seems, among other features, to have been wholly overlooked in distinguishing species of the wehhi and valens type. In valens the prothorax is undoubtedly very much shorter and more inflated at the sides than in convexa, but the statement by LeConte that it is nearly three times as wide as long is doubtless far overdrawn.^ In ocularis the prothorax is more transverse than in convexa and is more ante- riorly dilated, the eyes and antennae also being notably different from those of convexa in prominence and color, but, as these points are not mentioned by LeConte in regard to valens, further observation will be necessary before deciding definitely the kinship of ocularis and ' This proportional width of the prothorax is very deceptive. I have recently measured with dividers the width in terms of length, in cases where the prothorax certainly seemed to be beyond doubt more than twice as wide as long, as in Spinthoptera ocularis, for example, and the result proves that, even in such cases, it comes far short of actually being even as much as twice as wide as long. The comparative statements given in the descriptions of the present paper, as well as all others of the writer, are not based upon instrumental measurements, but are simply the ratios as they appear to his eye. The deception is due, in most part at least, to the shape of the apex, base and side margins, for if we had before us a geometric rectangle of the actual extreme dimensions, there would be much less trouble in realizing the true ratio. This goes to prove the necessity for stating the true dimensions of all the parts of an animal in linear units, as advocated by the writer many years ago, if we desire to record a description of the greatest possible value for purposos of comparison. STUDIES IN THE AMERICAN BUPRESTID^ 77 valens. It is distinctly stated by LeConte that the prothorax of webbi is wider at the base than at any other place, but Mr. Schaeffer seems to have mistaken this to be "wider at base than at apex," which of course is very evidently true of every species of the genus; I have assumed, however, that in cupreo punctata the greatest thoracic width is at or near the middle. I have been unable to identify insularis with any species of the Biologia and conclude therefore that it is as yet unrecorded in the literature of the subject; it seems to resemble hirtomaculata Hbst., somewhat, but that species is described by Gory as being slightly depressed, with the prothorax lightly pubescent later- ally. Chalcophora Solier. Although in this genus the general habitus is wholly unlike that of the preceding genera, the affinity may in reality not be so remote as appears, for the two prosternal sulci are present as in Spinlhoptera, though deeper and much more approximate. This and the following differ from any others of the genera here considered in having the epistoma more narrowed and much more abruptly and deeply emar- ginate, and the supra-antennal ridge, so prominent in the preceding genera, is here obsolete and replaced by a broad feeble rugose swelling of the surface. The antennae in Chalcophora are slender, with elong- ate and sparsely punctate outer joints, which appear to be wholly devoid of sensory fossae terminal or otherwise, being the only genus of this section of the family, coming under my observation, in which some trace at least of terminal fossae cannot be discerned; the terminal joint is evenly oval and without suspicion of apical process. The pro- notum is constantly broadly though feebly sulcate, the sulcus longi- tudinally divided by a slightly elevated obtuse smoother line, the scutellum small, the elytra gradually narrowed behind, entire at tip, not or but feebly serrulate at the sides posteriorly and with the sutural angle sometimes spiniform; their sculpture is peculiarly uneven, with large and finely punctate, more metallic depressions and polished lines and elevations, both having a generally though unevenly longitudinal arrangement. The legs are rather slender, the male tibiae without any striking sexual modifications, the first ventral impressed or sulcate and the basal joint of the hind tarsi at least as long as the two following combined. The male modification of the abdominal apex is more 78 CASEY pronounced than in any other genus except Texania, and consists of a deep, abrupt and angular emargination, not suggested at ail in the female. The male also has the median parts of the sterna more impressed and hairy than in the female as a rule, a character suggest- ing Buprestis and other following genera. The species are rather numerous and may be separated by the following characters: — Subsutural groove of the elytra abbreviated, attaining basal third or fourth; body more sombre in coloration, though similarly shining; basal sul- cus of the abdomen glabrous in both sexes 2 Subsutural groove continuing to the base of the elytra though shallow and sometimes subinterrupted near the base; lustre of the body more brightly aeneous or cupreous as a rule; basal sulcus of the abdomen, as well as the sterna, more pubescent in the male 8 2 — Sutural angles of the elytra distinctly dentate or spinulose; males not rare 3 Sutural angles not prolonged though sometimes feebly subprominent; males apparently much less abundant than the females 4 3 — Depressions of the elytra finely, densely and evenly punctulate, shining, blackish-aeneous, the depressions less shining and more cupreous; under surface obscure aeneous and partially cinereo-pubescent; head well developed, unevenly, moderately punctate, with a very large deep frontal impression which is narrowed posteriorly along the median line; eyes rather prominent; antennae not extending to the base of the prothorax even in the male; prothorax rather transverse, fully two- thirds wider than long, the sides broadly rounded anteriorly, sometimes prominent and occasionally distinctly angulate ( 9 ) , parallel toward base and generally nearly straight; surface broadly, roughly and un- evenly sculptured and biimpressed laterally, the two punctate sulci very feeble, the elevations smooth; scutellum small; elytra scarcely more than three-fourths longer than wide, the sides parallel, obliquely and arcuately converging in apical two-fifths and more or less visibly serrulate, the subsutural groove rather feebly concave, the second con- stricted and subinterrupted at and behind the middle, the sublateral carina nearly entire, the next distinct before the middle, uniting with the submarginal behind the humeri; under surface smooth, with small patches of fine close j)unctures. Length (ci') 20.0-22.5, (9) 25.0-28.5 mm.; width (0^)6.9-7.5, (9) 8.6-9.7 mm. Atlantic region, — Massa- chusetts to Florida. [=nov(B-boracensis Fitch, Buprestis virginiensis Drury, virginica Say and Harris, virginiaca Gmel., and cu preomaculata Goezej virginiensis Drury A — Nearly similar to virginiensis but more elongate and with the second elytral sulcus more interrupted at the middle, blacker in color, very shining, the under surface duller cupreo-icneous; head nearly similar but more coarsely, sparsely punctate, the eyes less convex; prothorax much less transverse, but little more than one-half wider than long, STUDIES IN THE AMERICAN BUPRESTID^ 79 the sides almost evenly and broadly arcuate, gradually converging anteriorly, parallel and straight basally, the surface rather more deeply impressed at the sides behind the middle, the median sulci extremely shallow and obsolescent; elytra similar but notably more elongate, the gradually converging and evenly arcuate sides poste- riorly even more evidently serrulate; basal sulcus of the abdomen extending beyond the middle of the second segment. Length (c?) 27.0 mm. ; width 8.5 mm. Florida obliterata n. subsp. Depressions of the elytra more coarsely and subrugosely punctate, the head less developed; body nearly as in virginiensis but with relatively smaller head and prothorax and rather more coppery lustre; head al- most similarly sculptured, the excavation rather narrower, the eyes a little less prominent; prothorax nearly similar in outline and sculpture but somewhat more transverse, the sides broadly rounded and con- verging anteriorly ( c?) or more subangulate anteriorly ( 9 ) , parallel and nearly straight basally; elytra much more evidently wider than the prothorax, otherwise nearly similar; under surface more coppery in lustre. Length (d^) 21.0-22.0 mm.; width 7.2-7.7 mm. Wisconsin (Bayfield) , — H. F. Wickham lacustris Lee. A — Nearly as in lacustris but with the head ( J') relatively still smaller and with a much narrower and feebler frontal excavation along the middle; prothorax still more transverse, swollen and rounded at the sides anteriorly, the sides parallel but broadly sinuate basally, the basal angles more acute; surface almost similarly sculptured; elytra less decidedly wider than the prothorax but otherwise nearly similar, the coppery depressions still more coarsely punctato-rugose and very much more strikingly so than in virginiensis; under surface nearly similar, cupreo-aeneous. Length (cJ') 22.0 mm.; width 7.8 mm. Massachusetts brevicollis n. subsp. 4 — First ventral scarcely more than flattened medially. Body very elongate, moderately convex, shining, black, the elytral depressions faintly bronzed, beneath feebly cupreous in lustre; head moderate, coarsely, irregularly and not closely punctate, the excavation broad anteriorly, narrow and deep along the median line posteriorly; eyes large but unusually feebly convex; prothorax two-thirds wider than long, the sides broadly rounded and gradually convergent anteriorly, parallel and somewhat sinuate basally, the basal angles unusually prominent; surface with the usual sculpture, the median impression very shallow, with the dividing ridge unusually narrow; elytra evidently wider than the prothorax, nearly twice as long as wide, the sides posteriorly unu- sually gradually convergent, becoming nearly straight before the broadly rounded apical angles, the sutural angle sharp and a little less than right, not at all produced; surface sculptured nearly as in virginiensis; under surface nearly as in that species but with the pubescence every- where very short and inconspicuous, the coarse punctures not so large. Length (9) 27.0 mm.; width 8.3 mm. Honduras. *hondurasica n. sp. First ventral distinctly sulcate along the middle as usual; sides of the pro- Proc. Wash. Acad. Sci., April, 1909. 8o CASEY thorax anteriorly more or less distinctly angulate or prominently and narrowly rounded, especially in the female 5 5 — Elytral depressions not or but slightly rugose, having fine dense punctures nearly as in virginiensis 6 Elytral depressions with larger, somewhat separated punctures, giving a much more coarsely rugulose appearance 7 6 — Elytra posteriorly more gradually obliquely narrowed, with the sides only very feebly arcuate; females only at hand. Body very elongate, moderately convex, shining, black, the depressions and the under sur- face bright cupreous; head well developed, irregularly punctate; an area along the eyes, much wider than in hondnrasica and a little wider than in virginiensis, finely, densely sculptured, the frontal pit deep and nearly as in the latter; eyes smaller and still more remotely sep- arated; prothorax nearly as in virginiensis, except that the median densely punctate impression is broader and somewhat shallower and the dividing median Hne very much wider and less convex; elytra nearly as in virginiensis but much more elongate, about twice as long as wide, the sides posteriorly with more obsolete or scarcely traceable serrula- tion, the apical angles broadly rounded, thence narrowly subtransverse to the suture; under surface with the usual moderately coarse, sparse punctures, with lateral patches of fine, dense and more metallic punc- tuation, the ashy hairs not conspicuous though rather long. Length (9) 26.5-28.5 mm.; width 9.0-9.8 mm. California. [ = Bnprestis angulicollis Lee] angulicoUis Lee. A — Nearly as in angulicollis but rather narrower and more convex, more sombre in coloration, the depressions dull subcupreous or bronzed, the under surface black, with feebly cupreous and finely punctured areas; prothorax less transverse, with less obvious anterior angula- tion at the sides, as in angulicollis very slightly wider at the angula- tion than at base, with the intervening sides tending to bisinuation; elytra nearly similar but with the sculpture of the depressions evi- dently less fine. Length ($) 24.5-26.5 mm.; width 8.5-9.0 mm. Colorado (Boulder Co.) montana n. subsp. Elytra posteriorly more rapidly converging and rounded at the sides, or, evidently more obtuse. Body somewhat broader in form and rather less convex, shining, black, the dej)ressions of the upper surface bronzed, the lower surface less brightly cupreous; head nearly similar, the eyes not prominent; prothorax more transverse, much more abru]>tly prom- inent at the sides at apical third (9) or not very markedly so (c?), the general surface more flattened basally, the sculpture nearly similar; elytra similar in general sculpture but with the raised polished spaces larger and less abruptly defmed and the sides more deeply sinuate at about two-fifths from the base, much less than twice as long as wide; metasternum more concave in both sexes, with unusually long and coarse white hairs in the male. Length (c?*, 9) 24.0-28.0 mm.; width 8.5-10.0 mm. Washington State and Idaho (Fort C(uur d'Alene) oregonensis Fitch STUDIES IN THE AMERICAN BUPRESTIDiE 8l 7 — Body broad and only moderately convex, shining, black, the depressions obscurely cupreous, beneath dull cupreous; polished black areas with feeble cupreous reflection; head as in virginicnsis, except that the an- tenna" are shorter and thicker than in the female of that species, the outer joints much less elongate; prothorax more transverse, very prom- inent laterally, and much wider than at base at about apical third, the sides thence converging and bisinuate to the base; surface sculptured in general as in that species, the depressions much more coarsely rugose and the median smooth line broader, the basal callus at lateral fourth much larger and more conspicuous; elytra broader, the converging sides posteriorly more rounded, with scarcely a trace of serrulation, the sculpture having the same general arrangement but with the elevations feebler and less sharply defined ; under surface with the usual sculpture. Length ( 9) 28.0 mm.; width lo.omm. Florida prominensn. sp. 8 — Smooth polished areas of the elytra but feebly elevated and less abruptly formed, in some parts dilated, the general sculpture nearly as in vir- giniensis 9 Smooth areas very abruptly formed, higher and linear in form throughout; sutural angle of the elytra never prolonged or dentiform; antennae thicker; elytra very evidently wider than the prothorax; species more boreal in habitat 10 9 — Form narrower and more elongate than in virginiensis, moderately con- vex, polished, generally with strong golden-aeneous lustre throughout above and beneath but occasionally black and without much metallic lustre; head well developed, unevenly punctured and rugose, the deep frontal sulcus not dilated anteriorly; eyes moderate, slightly prominent, the antenna slender; prothorax but little more than one-half wider than long, the sides parallel and nearly straight, gradually broadly rounding and very moderately converging anteriorly; surface sculp- ture as usual, except that the median smooth line is very broad and feebly convex; elytra about twice as long as wide, but little wider than the prothorax, the sides very gradually converging and feebly arcuate posteriorly, the apex unusually narrow, with the sutural angles strongly prolonged and spiculiform; surface with the smooth convex parts to- gether much exceeding in area the depressions, the latter finely and for the most part densely punctate; sutural sulcus becoming subobliterated basally; under surface with the pubescence of the finely and densely punctate areas very distinct, ashy white. Length (c?) 22.7-24.0, ( 9) 24.0-28.0 mm.; width (d^) 7.5-8.2, ( 9) 7-7-9-2 mm. Southern Atlan- tic States. [ = Buprestis georgiana Lee. , and liberta Lap.-Gory nee Germ.]. georgiana Lee. Form and sculpture nearly as in georgiana but narrower, the aeneous lustre varied with clouds of metallic bluish-green toward the sides of the elytra, the under surface bright subcupreous; head nearly similar, the eyes much less convex; prothorax trapezoidal, barely one-half wider than long, the sides distinctly converging and nearly straight from base to apex, very feebly arcuate anteriorly, the sculpture nearly similar; elytra inform and sculpture nearly as in georgiana but more evidently wider than the pro- 82 CASEY thorax and with the sutural angles at apex not in the least prolonged or dentiform, the apices very narrowly oblique; under surface nearly similar, except that the densely punctate areas are much less pubescent and the angular emargination of the last ventral in the male narrower, acute, with its sides straight and not of a bluntly ogival form as in that species. Length (c?) 22.5 mm.; width 7.0 mm. Southern Atlantic States iridescens n . sp. 10 — Head well developed, the eyes convex and rather prominent 11 Head notably small, the eyes continuing its generally rounded outline; body smaller in size than in any other species 13 II — Depressions of the elytra strongly, closely punctured and also rugose; elytra narrowing posteriorly from slightly behind the middle, with the sides evenly and very evidently arcuate to the moderately obtuse tips. Body elongate-suboval, more convex than usual, coppery in lustre, the elevations blackish, the under surface shining, cupreous; head densely punctate, also with some smooth rugae on its two large convexities, the latter separated by a broad and moderately deep groove; prothorax three-fifths wider than long, the sides parallel and somewhat sinuate, obtusely, subangularly rounded before the middle and thence strongly converging to the apex; sculpture nearly as in georgiana but vidth the sublateral impression behind the middle much shallower and more diffuse, the sublateral basal callus feeble ; elytra about twice as long as wide, notably convex, the sides feebly serrulate pos- teriorly, the apices rounded; under surface in great part finely, rather closely but not densely punctate, the pubescence inconspicuous, pol- ished and coarsely, sparsely punctate medially as usual. Length ( 9 ) 26.0 mm. ; width 9.0 mm. Pennsylvania forlis Lee. Depressions of the elytra rather more finely, densely punctate, not so rugose; elytra narrowing posteriorly in about apical third, with the oblique sides nearly straight toward tip, the apices transverse suturally 12 12 — Body rather broad, moderately convex, dull cupreous, the raised parts black, the under surface brilliant sneo-cupreous; head nearly as in fortis; prothorax relatively smaller, one-half wider than long, nearly similar in form and sculpture, except that the sides are less prominently rounded before the middle, sometimes very evenly rounded from behind the middle to the apex ; elytra evidently less than twice as long as wide, nearly as in fortis but less convex, the apex almost similarly obtuse; under surface nearly similar, except that the fine punctures laterally are denser, Length (c?) 23.0-25.0 mm.; width 8.0-8.8 mm. Can- ada (Ontario) laurentican. sp. Body broad, moderately convex, the depressions of the upper surface bright cupreous-red, the raised parts black; head and prothorax relatively smaller than in the preceding, the former nearly similar, the antennae notably thick, the subapical joints ( $) but little longer than wide; pro- thorax smaller, shorter and more transverse, three-fifths wider than long, nearly as in the two preceding, except that the parallel sides are apt to be somewhat bisinuate behind the apical converging and rounding parts STUDIES IN THE AMERICAN BUPRESTIDiE 83 and the sublateral impression and basal callus more pronounced; elytra distinctly less than twice as long as wide, much wider than the prothorax, the apex more acute than in either of the preceding, the sculpture similar; under surface with the sublateral areas of fine dense punctures smaller and more irregular among the smooth patches, and with the whitish pubescence much denser and more conspicuous, as usual decumbent but having also longer and more erect sparse hairs intermingled. Length (9) 21.0 mm.; width 7.7mm. Canada (east- ern Ontario). [=Buprestis liberta Germ., and borealis Lap. -Gory]. liberta Germ. 13 — Form somewhat as in liberta but narrower and more convex, the depressions duller cupreous-red, the elevations black, the under sur- face bright cupreous; head scarcely more than half as wide as the pro- thorax, less coarsely and more densely cupreo-punctate than in liberta, with narrower rugae, the sulcus similar; antennae (?) less stout, the subapical joints much longer than wide; prothorax similarly transverse, relatively still smaller, the sides parallel and obsoletely bisinuate, gradually strongly rounding and convergent from near the middle to the apex; sculpture nearly similar; elytra nearly similar throughout but still notably wider than the prothorax, the sides before the middle more deeply sinuous, the depressions similarly finely, densely punctate and nearly even; under surface with similarly isolated patches of dense fine punctures, which however are only feebly and very inconspicuously pubescent. Length (?) 18.0 mm.; width 6.8 mm. New York (north- em) parviceps n. sp. It will be observed that in the above statement some ten American species are considered valid, together with several subspecies, while, in the general catalogue of the Buprestidae, Captain Kerremans has listed only three, since fulleri belongs to Texania and not to Chal- cophora. This considerable increase of species, so material in fact that it may be criticized as unwarranted, is due principally to two causes: first to the fact that a number of species, long established, have been hastily and erroneously suppressed in the catalogue men- tioned, angulicollis and oregonensis, for example, having the elytra proportionally much more elongate than virginiensis and without the sutural spine of that species, being abundantly distinct and not synonyms as there enrolled. Then again a number of forms have been discovered since the genus was investigated by LeConte about fifty years ago, so that it is believed the actual increase of synonymy will at least not prove burdensome. The prothorax in Chalcophora is generally more prominent or subangulate at the sides before the middle in the female than in the male, as plainly observable in a series of virginiensis before me, in 84 CASEY some individuals of which the sides are not even notably prominent in anterior rotundity, while in others there is a distinct angulation; the character is, at best, rather inconstant and unreliable as a specific criterion. The synonym obscura Fitch, recorded originally by LeConte under virginiensis, and later repeated by Kerremans in the form '' obscurata,'' I have been unable to find in the quoted article by Dr. Fitch, There is, to be sure, a variety suggested by that author in the running text of his article, but it is not obscura, and, being unrecog- nizable, it were better that it lapse and be not perpetuated in synon- ymy. Texania n. gen. The more densely punctate antennal joints, forming one of the structural differences between this genus and Chalcophora, was noted without further comment by LeConte, but that author failed to ob- serve the large and well developed antero-internal sensory fossa of the outer joints, which are wholly obsolete in that genus; this, with the character mentioned, alone necessitates the generic separation from Chalcophora of Buprestis campesiris and other allied forms. In addition, however, there are three other distinctive structural char- acters of importance distinguishing the two genera. The first of these supplemental characters relates to the general sculpture of the upper surface of the body, the pronotum having a narrower, deeper and undi- vided median sulcus, and the more coarsely punctato-rugulose elytra have very shallow, irregularly rounded areolae, more or less interrupt- ing the continuity of a few fine uneven longitudinal nervures. The second differential character results from the complete coalescence or disappearance of the prosternal sulci of Chalcophora, to form a feeble pubescent impression in the male or simply a flattened surface in the female, and the third refers to the very strong and acute serration of the sides of the elytra posteriorly, which, however, is not always so developed as in the type, becoming much feebler, though always dis- tinct, in such species asfulleri and bisinuata. At the same time, the affinity of Texania with Chalcophora is strongly expressed by the general structure of the head and prothorax, as well as by the form of the triangular incisure at the abdominal apex of the male. Kerre- mans separated a genus Chalcophorella from the European representa- tives of Chalcophora, but, from all information available at present^ STUDIES IN THE AMERICAN BUPRESTIDiE 85 it cannot be identical with Texania, although seemingly a valid genus and not a subgenus of Chalcophora. The species known to me are not very numerous and may be dis- tinguished as follows: — Species of smaller size and narrower form, the pronotal sulcus narrow and deep, the serration of the elytral margins posteriorly very strong and conspicuous 2 Species of large size, broader and more depressed form, the pronotal sulcus broad and shallow, the elytral serration much feebler 3 2 — Body narrow, elongate, moderately convex, shining, with feeble obscure greenish lustre, nearly glabrous, the under surface brighter cupreous and with stiff, sparse and inconspicuous decumbent hairs; head rather small, the front nearly flat, coarsely and confusedly punctate and rugose, the median line finely grooved, with a small oval pit between the pos- terior parts of the eyes, the latter moderately prominent; antennae not attaining the thoracic base in either ses ; prothorax trapezoidal, from a little less (d^) to somewhat more (9) than one-half wider than long, the sides sensibly converging and feebly, subevenly arcuate throughout, the basal angles slightly everted, acute and prominent; surface with the sulcus entire, deep and acutely excavated, confusedly punctured and rugose broadly toward the sides, with a deep sublateral impression be- hind the middle, the basal margin very feebly sinuate at the scutellum, the latter very small, rounded; elytra slightly wider than the prothorax, a little more than twice as long as wide, parallel, the sides gradually rounding and converging behind in more than apical two-fifths, the conjoined apex feebly emarginate at the suture, the apices very narrowly sinuato-truncate; surface coarsely, feebly, unevenly punctato-rugose, with obscure longitudinal striae between the pronounced nervures* the areolae finely but sparsely punctate and shining; sterna and first ventral throughout deeply impressed (d^) or flattened to feebly con- cave (9), the last ventral in the former with a large equilatero-tri- angular emargination, the under surface of the last dorsal appearing beyond the incisure, flat, densely punctate, puberulent and with a finely and strongly beaded free edge; legs and coxae nearly as in Chal- cophora. Length 22.0-24.0 mm.; width 6.7-7.8 mm. Missouri, Kentucky and Indiana. [ = Buprestis campestris Say and substri- gosa Lap.-Gory] campestris Say Body slightly broader, similarly convex, rather less shining, greenish in lustre, with short hairs particularly evident posteriorly, the under sur- face cupreous, with brilliant green reflection along the median parts of the sterna, the pubescence more evident than in campestris, the punctures laterally, as in that species, small and subevenly distributed, not in dense patches as in Chalcophora; head slightly larger but other- wise nearly as in campestris, though more evidently and broadly con- cave toward the median line throughout; prothorax more transverse, the sides nearly parallel for three-fifths from the base, there broadly 86 CASEY rounded and thence strongly converging and nearly straight to the apex, the median sulcus wider, less acute at the bottom, the sublateral impression much more feeble and more diffuse; elytra nearly similar in form and sculpture but more densely and rugosely punctate through- out, especially in the areolae, the apices narrowly but more obliquely truncate, the sutural angle prominent and acute; male sexual charac- ters similar; tarsi more metallic green in lustre. Length ( c?) 23.0 mm. ; width 7.2 mm. Pennsylvania serriger n. sp. 3 — Form oblong, depressed, dark bronze above, feebly shining and sparsely pubescent in patches; head coarsely punctate, deeply, longitudinally grooved; prothorax broader than long, the sides obtusely angulate med- ially, oblique anteriorly, subparallel behind, feebly sinuate before the hind angles, which are moderately prominent; apex feebly emarginate, the base bisinuate; surface with a broad vague channel medially and a moderately deep longitudinal impression sublaterally, coarsely rugulose at the sides, smoother medially; elytra oblong, depressed, parallel, gradually narrowed and moderately serrate at the sides in apical third, the apex obtuse; surface with four vague nervures, the latter with thickened spaces which surround large but very vague fovese [areola?], the nervures feebly shining, the intervals opaque, rugulose, sparsely punctured and with very short pubescence; under surface similar in color to the upper, irregularly and coarsely punctate, with spots of denser and finer punctuation. Length 27.5 mm. Texas. [= Chalcophora fulleri Horn] fuller! Horn Form stout, very moderately convex and not very shining, with a pronounced metallic green lustre above, each puncture with an extremely small hair, which is more developed in the male; under surface cupreous and polished along the middle, greenish and duller laterally, the punctures fine and dense laterally, with numerous small smooth patches, the pubescence extremely minute and inconspicuous ( 9 ) or longer, dense and very distinct ( d^) ; head moderate, densely and coarsely punctato- rugose, narrowly excavated along the median line, with a small deep fovea nearly as in campestris; antenna; apparently longer in the female than in the male but not attaining the thoracic base; eyes moderately prominent; prothorax fully two-thirds wider than long, the sides par- allel and evenly bisinuate to well before the middle, there obtusely angulate and thence strongly convergent and straight to the apex; base but feebly impressed at the scutellum; surface finely, sparsely punctate medially, coarsely and rugosely laterally, the median sulcus very broad, deeper and more densely punctulate anteriorly, becoming gradu- ally obsolete basally, with the median line deeper just before the basal margin, the large sublateral impression, just behind the middle, feeble and indefinite; elytra about twice as long as wide, evidently wider than the prothorax, parallel, the sides rapidly converging, arcuate and slightly serrate in apical third, the apices cupreous, narrowly oblique, with the sutural angle produced and acute; surface somewhat finely, closely punctato-rugose throughout and similarly so and equally shining and puberulent in and outside of the areola;, the nervures not as prominent STUDIES IN THE AMERICAN BUPRESTID^ 87 as in campestris, the sexual characters of the under surface nearly sim- lar. Length (c?, 9) 25.0-30.0 mm.; width 8.5-10.8 mm. Texas. bisinuata n. sp. Form broader and less parallel than in caw/><75/m, the size larger, the pronotal sulcus broader and more shallow, visible for only three-fourths of the length and having a slightly flattened elevation near the anterior mar- gin; prothorax more " triangular, " the elytra more dilated. Louisiana (near New Orleans). [= Chalcophora langeri Chev.]. . langeri Chev. The species langeri Chev., is of rather uncertain tenure; the few lines of description given above are drawn from information furnished LeConte by Chevrolat himself and printed in the Monograph of the former author. It apparently cannot be the species described above under the name bisinuata, neither does it seem to be the fulleri of Horn. The description of the prothorax as triangular is rather indef- inite, fitting campestris better than any other, but the characterization of the thoracic sulcus will not at all apply to that species.^ I have therefore appended it provisionally as above and until further material from Louisiana can be collected, for as yet we certainly do not know all concerning the fauna of that section of the country. There are probably a number of species of Texania in Texas and territory immediately adjacent, which region appears to be its proper centre of dispersal. All the species of Texania seem to be more or less rare in collections. Buprestis Linn. Ancylodieira Esch.; Anoplis Kirby. The species of this genus are properly holarctic in range, some trop- ical and Australian forms now attached belonging in reality to other genera; they are numerous and troublesome to separate with certitude, because of considerable variation in form, size and sculpture within specific limits; many of them also are rare or local. It is possible, however, by careful study of large series, to become sufficiently fam- iliar with the laws of intraspecific variation, to be tolerably sure in ^ Such evidence as we have shows beyond doubt that langeri is a species of the fulleri type, and that the probability is that the expression "prothorax more triangular" refers to the very patent fact that the sides of the prothorax are more angulate than in campestris. It may be therefore that either fulleri or bisinuata is a synonym of langeri. Or perhaps both fulleri, which I have not seen, and bisinuata may prove to be varietal forms or synonyms of langeri. 88 CASEY the taxonomic assignment of those forms represented by single exam- ples, although of course mistakes under these circumstances may readily occur. Difficulties of this kind, nevertheless, scarcely war- rant such carelessly sweeping opinions as that put forth by Crotch, in surmising that all the immaculate forms allied to maculativentris, constitute a single species, for this is certainly very far from being true. It is also most improbable that there can be the transitions be- tween lineata and maculipennis hinted at by LeConte, as apparently proved, at any rate, by the large series at hand. The genus Buprestis, as represented by octoguttata, which was sug- gested by Kerremans as the type to be definitively adopted and to which no valid objection can apparently be advanced, includes a wide range of bodily habitus, but the elytra are regularly striate throughout, without the faintest indication of such irregularities as the depressions or areolae of the preceding genera, and this would seem to be the most important general distinguishing character of Buprestis within this group having the outer antennal joints elongate, more narrowed toward base, inferiorly punctate and with a distinct antero-internal sensory fossa. The frontal margin is always broadly sinuato-truncate, the antennas inserted in triangular cavities which are generally accentuated internally by the moderate though clearly de- fined obhque ridges; the latter, however, become obsolete in langi, rufipes and related forms. The labrum is usually wholly pale and coriaceous, but in langi and allies it becomes partially corneous and metallic in lustre; it seemed at first as though this disparity in labral structure might form the basis for subgeneric division, but as it would have placed two such very similar species as rufipes and elongala in different subgenera, it evidently fails to have any such value. The pronotum generally has a smooth and punctureless, more or less embossed median line, but in langi and other similar forms it is wholly devoid of this character, the surface being feebly impressed and more punctured along the median line. The prosternum and first ventral segment may be impressed to strongly sulcatc, or without trace of impression and with pronounced even convexity. The legs are slen- der, the basal joint of the hind tarsi more or less considerably elongate, and the anterior tibiai of the male may have a strong reflcxed spine internally near the tip or betray no vestige of any such structure. The abdominal apex is more or less truncate in both sexes, though, as STUDIES IN THE AMERICAN BUPRESTID^ 89 in conjluenta, sometimes more deeply sinuate in the male, but these secondary characters are very inconstant and liable to mislead, so that they are generally not referred to in the descriptions. The integu- ments are always virtually glabrous, shining and frequently metallic in coloration, though sometimes deep black, some species being among the more lugubrious of this highly metallic family; some are maculate with flavate in diverse patterns and others unicolorous. The species and subspecies represented in my collection, or assign- able with some degree of certainty from published descriptions, may be identified by the following tabular statement : — Male with the anterior tibiae deeply sinuate within beyond the middle and with a strong reflexed subapical spine internally; pronotum never impressed along the median line 2 Male with the anterior tibiae unmodified; pronotum frequently impressed along the middle 29 2 — Presternum flattened, sometimes feebly impressed along the median line, the first ventral segment always sulcate medially 3 Prosternum broadly, evenly convex, the first ventral never more than feebly impressed along the middle; elytra with finely comminuted macula- tion, the striae and intervals subuniform throughout the width 28 3 — Elytra each with two brick-red vittae, sometimes more or less disinte- grated, the inner much abbreviated. Body elongate-fusiform, con- vex, black, generally with feebly aeneous or cupreous reflection, less marked on the elytra; head about as wide as the thoracic apex, rather finely, densely punctate, less densely and more coarsely anteriorly and in a posterior interocular tumescent transverse area, having (c?) an anterior arcuate and irregular and a post-median transverse pale fascia, or ( 9 ) usually two or four small apical and two juxta-ocular spots; antennae slender; prothorax nearly three-fourths wider than long, strongly trapezoidal, the sides nearly straight, sometimes slightly in- flated basally, generally linearly pale at the lateral and apical margins toward the angles, rather coarsely, subevenly and not very densely punctate, with a slight impression before the scutellum; elytra but little wider than the prothorax, three-fourths longer than wide, the sides gradually rounding and oblique in posterior two-fifths, the apices trans- versely truncate with spiculate inner and outer angles, the truncature also feebly and irregularly microspiculose; striae moderate, finely punc- tate, the intervals but feebly convex, more coarsely punctate, almost uni- form throughout, the fourth and especially the sixth notably broader basally and more punctate; under surface shining, distinctly punctate, the last ventral with a smafl red spot near each anterior angle. Length 1 2. 2-1 7.0 mm.; width 4.8-6.7 mm. North Carolina, Florida and Louisiana lineata Fabr. Elytra each with four flavate spots arranged longitudinally and subequally 90 CASEY spaced, the spots not interrupted in continuity by the striae, though sometimes divided and forming two or three spots, one or all of which may become obsolete, the two or three anterior spots frequently united between themselves longitudinally and internally 4 Elytra each with four large flavate spots or narrower bands, which are dis- continuous, being interrupted at the stride; abdomen with four series of pale spots, generally obsolete in the males 11 Elytra immaculate 16 4 — Body smaller, more fusiform, the prothorax subevenly trapezoidal and almost evenly punctate, nearly as in lineata, the elytral apices truncate, feebly spiculose and bidentate, the strial intervals nearly flat and almost equal. Atlantic regions 5 Body larger, more elongate, the prothorax trapezoidal but strongly swollen at the sides basally and with more irregular sculpture; elytra evenly truncate at tip, with the outer angle obtuse to feebly dentate; elytral intervals alternately strongly convex. Western North America 9 5 — Second and third spots of the elytra when present always strongly and outwardly diverging, approximate and generally united at their bases. 6 Second and third elytral spots nearly transverse, widely separated longitudi- nally even internally, the spots generally large, fully formed and not at all comminuted 8 6 — Larger as a rule or stouter and with a larger head, more southern in habitat, the ante-coxal plate of the metasternum generally bimaculate medially. Rather stout, convex, shining, black with feeble aeneous lustre; head (9) well developed, rather coarsely, somewhat unevenly punctate, the eyes moderate, widely separated, bordered internally with orange-red, which becomes transverse superiorly, the front also with two anterior spots and a central detached vitta of the same color; antennse slender, not attaining the thoracic base; prothorax three- fifths wider than long, evenly trapezoidal with straight sides, the apex transverse, broadly arcuate medially, the base bisinuateas usual; punc- tures rather coarse, unevenly sparse medially, closer laterally; scutel- lum rounded, flat, polished and metalHc; elytra less than twice as long as vAde, the sides gradually rounded and oblique posteriorly, though feebly arcuate, to the apical truncature, the striae strong, closely punc- tate, the intervals more coarsely punctate, nearly flat, alternately nar- rower with a single series of punctures and slightly broader with the punctures somewhat closer and more confused; flavate spots small and much disintegrated; abdomen coarsely, closely and conspicuously punctate, the basal sulcus rather deep, continuing virtually to the seg- mental apex, the last segment with a transverse orange-red spot at each side near the base. Length (9) 15.0 mm.; width 5. 85 mm. Florida. maculipcnnis Lap.-Gory A — Similar to maculipennis but narrower and relatively more elongate, the head ( 9 ) fully as developed, with rather prominent eyes, the lower flavate margin along the latter interru])ted at the middle and not extending upward as far as the middle of the eyes, also with two STUDIES IN THE AMERICAN BUPRESTID.E QI small frontal spots, the posterior spots seldom and only sporadically indicated, the antero-inferior thoracic and two anterior marginal trans- verse spots of the prosternum less developed; elytra narrower, with similarly small, sparse and very broken pale maculation; under surface with bright metallic lustre, the abdominal punctures rather strong but sparser, the orange-red spot at each side of the last segment more posteriorly oblique inwardly, sometimes almost obsolete. Length (9) 1 2. 7-13.5 mm.; width 4.8-5.4 mm. North Carolina (Southern Pines) deflciens n. subsp. B — Form shorter, stouter and less convex, not so shining or metallic above, the under surface dark subcupreous; flavate border of the eyes ( $ ) extending upward to the middle of their inner margins, with- out other pale frontal marking in the type; prothorax nearly as in nmculipennis, the elytra shorter and broader, two-thirds longer than wide, the flavate spots large, solid and complete, the first and second united, the external part of the latter curving forward externally, enveloping the humeral callus and virtually attaining the base, the third touching the second internally and similarly strongly though posteriorly oblique, the fourth small and subapical; abdomen nearly as in the two preceding but less strongly punctate, the lateral spot of the last segment obsolescent. Length ( 9 ) 14.0 mm. ; width 5.9 mm. Virginia (Cape Henry) fusiformis n. subsp. Smaller and more boreal in habitat, the head notably smaller in size, the ante-coxal plate of the metasternum apparently very seldom maculate with paler tint 7 7 — Abdomen narrowly and rather deeply sulcate at base as usual, with smaller, sparser punctuation; pronotum without well defined and more convex impunctate areas. Body narrowly fusoid, convex,^ shining, feebly subsneous; head rather small, the eyes not very prominent, the lower yellow stripe along the eyes terminating at their lower third ( 9 ) or with this border ascending to the upper part of the eyes, bisinuate within and transversely, inwardly prolonged at its upper end, with a large anterior anchor-shaped spot ( c?) ; prothorax nearly as in deficiens but more transverse, the elytra nearly as in macidipennis , with the flavate spots more or less disintegrated; abdomen with the transverse lateral spot frequently wholly obsolete. Length (d^, 9) 9.5.-14-0 mm.; width 3.8-5.5 mm. Massachusetts, Pennsylvania (Hazleton) and Indiana inconstans Mels. A— Form more evenly fusoid, the head ( 9 ) still rather smaller, with the inferior yellow line along the eyes and two small anterior spots; pro- thorax similar, except that the sides are more converging from base to apex; elytra similarly striate but with the sides posteriorly more evenly arcuate to the truncature and not straight or even feebly sinuate, as they generally are in inconstans; maculation more com- plete, nearly as m fusiformis; abdomen more finely and feebly punc- tate, the ante-coxal plate of the metasternum feebly bimaculate in the type. Length ( 9 ) 12.2 mm.; width 5.0 mm. Michigan. gcripta n. subsp. 92 CASEY Abdomen at base narrowly, very feebly impressed along the flattened median line and coarsely, very densely and conspicuously punctured throughout; pronotum with some distinct impunctate callous areas. Body very small, narrowly fusoid and convex, colored and sculptured otherwise as in inconstans, the head ( 9 ) very small, with the eyes not prominent, having a small pale spot adjoining the lower limb of the eyes; protho- rax shorter and more transverse, at base nearly twice as wide as long, the ante-scutellar impression unusually deep and very distinct; elytra as in inconstans, the sides arcuate posteriorly to the truncature; last ventral without trace of maculation in the type. Length ( 9 ) 10.7 mm.; width 4.2 mm. New York (Adirondack Mts.) reducta n. sp. 8 — More elongate, fusiform and less convex than in the preceding forms, the maculation orange-red; head (9) as in inconstans, or (d^), with the entire anterior part solidly pale, excepting two small black points, the pale area deeply, angularly biemarginate posteriorly; prothorax nearly to quite twice as wide as long, almost as in inconstans; elytra gradually becoming oblique posteriorly but with the sides feebly arcu- ate to the truncature, striate nearly as in inconstans, but with the punc- tures of the intervals less coarse, the second interval from the side mar- gins more prominently convex; abdomen sulcate as usual at base, not very coarsely but strongly, moderately closely punctate, the spots of the last segment very strongly, inwardly oblique. Length ( J^, 9 ) 1 1 -o- 14.5 mm.; width 4.2-5.6 mm. Michigan and Missouri. leporina n. sp. 9 — Body black-bronzed, polished; under surface with a few pale decumbent hairs; head confluently punctured, with several irregular confluent smooth spaces; labial palpi, spot on the mandibles, labrum, lower margin of the eyes and some frontal spots yellow; prothorax bisinuate at apex and base, coarsely punctate, with several smooth spaces, the lateral margin except basally, and part of the apical margin yellow; elytra slightly furrowed, the furrows punctured, the intervals alter- nately convex and plane, the sutural interval convex, forked at base, the flat intervals most punctured but the convex the more coarsely; surface with three equidistant and irregular yellow spots, arranged longitudinally, and, nearer the base on the second ridge, there is a line of confluent yellow dots; apices truncate; abdomen having a triangular orange-colored spot at the sides of each segment, those on the last seg- ment larger and irregular, the coxae also and under sides of the thighs are partly of the same color. Length 15.0-16.0 mm. British- Amer- ica (Latitude 65° and in the Rocky Mountains) nuttalH Kirby Body black; abdomen with two widely separated series of small pale spots at each side; coxae and femora only extremely rarely exhibiting any paler maculation lO 10 — Form oblong-oval, moderately convex, deep black, without metallic coloration above, feebly metallic beneath; head wholly bright orange, with two rounded central black spots arranged transversely, the pos- terior line of the pale area, between the upper limits of the eyes, with a large median transverse sinus, the punctures generally well separated STUDIES IN THE AMERICAN BUPRESTIDiE 93 (c5^), or more densely and irregularly punctate, black, with orange maculation along the eyes and with two or three small anterior spots ( ? ) ; prothorax trapezoidal, swollen laterally toward base, three-fourths wider than long, the anterior and lateral margins near the apical angles pale (cJ*), or with merely a spot under the angles pale (9); surface rather coarsely, sparsely punctate, with several large irregular smooth areas sublaterally and a broadly smooth median line, fmely striate subbasally; elytra three-fourths longer than wide, the sides parallel and nearly straight for two-thirds, thence oblique, becoming nearly straight to the apices, which are simply transversely truncate (9), or obliquely so with the sutural angles more prominent (c?); surface with deep but finely and very obscurely punctate striae, the intervals strongly convex and very sparsely, not coarsely punctate, alternately only slightly more so, except the sixth and eighth which are decidedly less convex and more punctate though not densely, the former greatly broadening over the humeral swelling, the reddish-flavate spots two and three alone well developed, vaguely subquadrate and feebly, divaricately oblique outwardly, the first spot small, more internal and generally comminuted, the fourth small and sometimes obsolete; under surface sparsely pubescent, variably punctate, the abdomen deeply sulcate basally, the four spots of the fourth segment subequidistant (9), or vidth the two median very approximate (c?). Length 16.5-18.5 mm. width 6.4-6.9 mm. New Mexico (unrecorded locality and Fort Win- ga^te) alternans Lee. A — Form nearly as in alternans but much larger, with the elytra more gradually narrowed and more produced posteriorly, the eyes rather less prominent, the prothorax more inflated and rounded laterally in nearly basal half, the sides thence more converging and straighter to the apex. Length (9) 20.0 mm.; width 7.7 mm. Arizona (Walnut) conicicauda n. subsp. Form much more cuneate and more elongate, similarly convex, the upper surface generally with a faint greenish-metallic lustre; head (d", 9) respectively as in alternans, except that the transverse posterior margin of the solid bJ'iigronotate pale area of the former has a very small deep median sinus; antennas (cT) scarcely attaining the thoracic base; eyes but slightly prominent; prothorax less transverse but otherwise nearly similar, with a much stronger impression before the scutellum; elytra generally more sinuate at the sides near the middle, gradually oblique posteriorly and slightly arcuate to the apical truncature, the outer angle more obtuse, blunt; surface nearly as in alternans but with the reddish- flavate pale spots much more developed, all four distinct, the first and second generally united and sometimes also the outwardly and divari- cately but notably more strongly oblique second and third spots; ab- domen deeply sulcate at base, finely and sparsely to rather strongly and more closely punctate, the pale spots frequently, and especially in the male, wholly obsolete, excepting two rather large and irregular on the last segment. Length 16.7-18.7 mm.; width 5.9-7.0 mm. California to Washington State. [^Ancylocheira (/z7a/a/a Motsch., i litt. ?]. laevivcntris Lee. 94 CASEY A — Similar to IcBviventris, except that the abdominal surface is not pol- ished but feebly alutaceous, the hind coxae not uniformly black but with a transverse pale discal spot, and the large spots of the last seg- ment not full and vaguely triangular but transverse, each with a slen- der vitta proceeding from the side posteriorly and extending parallel to the sides almost to the tip; prothorax more strongly and abruptly swollen at the sides in basal half. Length (?) 21.0 mm.; width 8.0 mm. Washington State pugetana n. subsp. II — Pronotum strongly impressed near each side just behind the middle. Body oblong-elongate, rather feebly convex, black, without distinct metallic lustre, the under surface with short sparse hairs; head densely punctato-rugulose, sparsely pubescent, with a narrow elongate callus medially and a fine stria thence to the base, the eyes moderate; front with anterior as well as juxta-ocular pale spots in both sexes; antennae unusually thick; prothorax very nearly twice as wide as long, the sides inflated and strongly, evenly rounded in basal half, thence straight and converging to the apex, the latter and base strongly bisinuate; surface with the margins near the apical angles and a small elongate spot near the basal pale, the punctures coarse, close, dense and rugulose laterally, the four smooth vittse very pronounced, the impression within as well as without the sublateral irregular vittae distinct, the median vitta strongly striate along the middle except apically, the two anterior callous spots large and well defined; elytra parallel, the sides gradually rounding posteriorly to the wholly unextended apices, which are trun- cate and usually bidenticulate, the surface with the alternate intervals perfectly flat, subopaque, finely and densely punctate, or very convex, polished, not more coarsely but very sparsely punctate, the four pale fasciae broken into small elongate spots on the ridges; under surface with small pale spots on the prosternum, met-episternum, coxae and along the under side of the femora; abdomen strongly punctate, sulcata at base, wholly immaculate (J*), or with a series of very small red dots along each lateral margin ( $). Length 17.5-20.0 mm.; width 6.8-7.7 mm. New Mexico diruptans n. sp. Pronotum not or feebly impressed sublaterally, the sides strongly swollen basally, thence converging and almost straight to the apex; elytra with the intervals equally polished throughout, the flatter and more convex not so unequal and sparsely, subsimilarly punctate, the pale fasciae more developed, attenuate outwardly and interrupted only at the striae, the second tending to curve forward anteriorly at the sides; apices always truncate, bispiculose and frequently very feebly, irregularly undulate; under surface with small pale spots on the sterna and coxae and sometimes on the under side of the femora 12 12 — Elytra unusually prominent at the sides near basal fourth and thence strongly sinuate to behind the middle; male with two small convexities in the apical margin of the second ventral, the median part of the seg- ment flattened, with the coarse dense punctures streaming obliquely toward the two inequalities. Body black, ^vithout metallic lustre; head (cJ') not coarsely but closely punctate, with a median and two STUDIES IN THE AMERICAN BUPRESTIDiE 95 rather distinct oblique smooth lines, the median line of the occiput broadly smooth and concave; anterior half of the front yellow, with a posterior spur at the middle and two small anterior black points, also with a more j^osterior oblique spot near each eye, the latter large but not prominent; antennae rather slender; prothorax small, moderately transverse, the sides arcuate basally, the apex broadly prominent and arcuate, only very feebly sinuate laterally; surface coarsely, closely punc- tate, with pale margins and impunctate areas as in dimplans, the sub- lateral smooth vitta partially interrupted by a small deep foveiform impression; elytra much wider than the prothorax, strongly cuneate from the base, broadly rounded and subprominent laterally behind the middle, the sides thence rapidly obUque, becoming subsinuate to the apex; under surface strongly punctate; femora and abdomen without pale spots, except a small one at each side of the last segment. Length ( c?) 1 7.0 mm. ; v^idth 6.3 mm. Washington State (Tacoma). contorta n. sp. Elytra subparallel, not very prominent subbasally and but feebly sinuate laterally, the second ventral unmodified in the male 13 13— Basal inflated part of the prothorax with the sides broadly rounded. Pacific regions 14 Basal inflation more abrupt, angulate or tumescent, the head, pronotum and under surface more cupreous. Atlantic regions 15 14 — Form oblong, parallel, only moderately convex, shining, colored as in contorta and with the general characters similar, except that the head and prothorax are relatively larger ( c?), the latter almost as wide as the elytra, the head nearly similar in sculpture and ornamentation, except that the upper pale spots near the eyes are less oblique and less poste- rior; eyes rather small and not at all prominent; prothorax with the apical margin even more strongly and prominently arcuate medially, the callous smooth areas similar, the small impression on the outer side of the sublateral vitta less marked; sides near the base less broadly prominent than in the two following forms; elytra but three-fourths longer than wide, very rapidly oblique behind in less than apical third; under surface strongly but not densely punctate, shining, with the usual sternal and coxal pale spots, the femora with an elongate pale spot beneath, the abdomen with a pale spot at each side of the basal sulcus and another transverse at each side of the last segment. Length (c?) 17.8 mm.; vddth 7.4 mm. California gravidula n. sp. A— Nearly as in gravidula throughout though a little narrower, the head ( d^) with larger and much more prominent eyes, the median line and two lateral discal spots smooth, the pale area confined to the entire apical margin, running along the eyes from above the middle to the base of the head beneath, the median posteriad pale spur long; prothorax similar; elytra more elongate, more gradually and less strongly oblique at the sides posteriorly in apical two-fifths, feebly arcuate even near the apices; under surface nearly similar, the femora and abdomen without pale spots, except a slender oblique line at each side of the Proc. Wash. Acad. Sci., April, 1909. g6 CASEY last segment; basal joint of the hind tarsi long as usual in this group, almost as long as the next three combined. Length (6^) 17.8 mm,; width 6.8 mm. Washington State torva n. subsp. B — Nearly as in torva but still more elongate, the elytra nearly twice as long as wide, with the sides similarly gradually rounding and oblique in posterior two-fifths but becoming nearly straight to very feebly sinuate near the apices; head (?) as in gravidula, having the anterior margin throughout the width, sometimes broken into four spots, pale; prothorax as in that form but with feebler indication of impres- sion interrupting the sublateral smooth vittas; elytra sculptured as in torva, the flatter intervals rather more punctate than in gravidula; abdomen with the four lines of pale spots well developed (9), with numerous pale spots on the sterna and coxae and one on the under side of the femora, the spots of the two inner abdominal lines larger than those of the marginal lines. Length ($) 17. 3-19. 7 mm.; width 6.2-7.8 mm. Colorado (Boulder Co.). . boiilderensis n. subsp. 15 — Form nearly as in the preceding, the size much smaller, the coloration and sculpture almost similar; head (c?) densely punctate, the callous areas very small and inconspicuous, the occiput impressed along the median line; front yellow in anterior half, except a sinus above each antenna and generally two rounded discal points, the yellow area with a short posteriad median prolongation, the spot near the upper part of each eye distinct to wholly obsolete; eyes somewhat prominent; prothorax with the sides at two-fifths from the base subangularly pro- tuberant, the apical margin broadly bisinuate; sculpture and orna- mentation as in gravidtda, except that the impression near the outer smooth vitta is wholly obsolete; elytra as in the preceding forms grad- ually oblique at the sides posteriorly, sometimes feebly sinuate toward the apices; under surface and sometimes the femora inferiorly spotted with flavate as in the preceding; abdomen strongly punctate, narrowly sulcate at base, generally with a pale spot at each side of the sulcus, the other spots nearly obsolete ( J'), the single spot at the centre of the fourth segment, forming the apex of the two inner series, generally visible, the last segment with a biramose spot at each side, which is however frequently wholly obsolete. Length (c?) 13. 5-16.7 mm.; ^vidth 5.0 6.0 mm. Lake Superior (Duluth and Bayfield) flavopicta n. sp. Form notably narrower and relatively somewhat more convex, the pale broken fasciae of the elytra larger, almost always longitudinally conflu- ent near the suture; general coloration and sculpture similar; head smaller, less densely punctate and not so evidently pubescent, similarly ornamented; prothorax similar but less transverse, the sides less angu- larly tumid and at basal third, thence to the base frequently with an intermediate sinus; elytra somewhat more produced at tip, with the sides toward the apices more evidently sinuate, the apical truncature transverse as usual, finely bidenticulate; under surface nearly similar but with less evident ashy hairs, the abdomen more sparsely and evi- dently less strongly punctate as a rule. Length (cf) 13. 5-1 4. 2 mm.; width 4.8-5.2 mm. Pennsylvania and New York to Michigan. consulari.s Lap. -Gory STUDIES IN THE AMERICAN BUPRESTID^ 97 i6 — Abdomen with four series of pale spots; body convex and fusiform, generally with rather distinct green or violaceous metallic lustre 17 Abdomen with a single series of pale spots at each side 21 17 — Pronotum^strongly declivous laterally, the side margins and apical angles not visible from above. Form elongate, convex, the lustre pol- ished, feebly violaceous, more evidently so beneath; head strongly and not very densely punctate, the median line narrowly smooth anteriorly but without other callous spots, the eyes ( $ ) margined internally throughout with an internally bisinuate red border, the front also with two anterior elongate red spots; eyes rather prominent; antennae mod- erate; prothorax three-fourths wider than long, unusually parallel, the medially subprominent apex fully three-fourths as wide as the base, the sides evenly converging from base to apex, evenly and feebly arcu- ate; surface evenly convex, with an oval orange spot at the sides of the apex, coarsely, generally sparsely and unevenly punctate, having sub- lateral areas and the median line impunctate, the latter wath a small deep basal puncture; elytra but little wider than the prothorax, twice as long as wide, parallel, the sides very gradually rounding and con- verging in apical two-fifths, the apices rounded, transverse toward the suture, the latter scarcely prolonged; surface wdth the intervals alter- nately wider, moderately convex and sparsely punctate and narrower, flatter and more closely punctate; under surface strongly punctate, the abdomen less densely, shining, wholly immaculate excepting two small rounded red spots at each side of segments two to four and one transverse near each side of the fifth, sulcate basally. Length ( 9 ) 17.8 mm.; width 6.8 mm. Colorado (?) — Cab. Levette. rubronotans n. sp. Pronotum not strongly declivous laterally, the side margins and apical angles visible from above 18 18 — Prothorax much narrower than the elytra, the sides not continuous. Body less convex, polished, the anterior parts feebly, the elytra strongly metallic green, the under surface violaceous; head (9) as in the pre- ceding but with the eyes less prominent and with more developed pale orange marking, the latter completely enclosing the front except at the middle posteriorly; prothorax two-thirds wider than long, trape- zoidal, the sides evenly converging throughout, very feebly arcuate ex- cept apically, the margins antero-laterally orange- red; surface coarsely, closely, unevenly punctate, tumid submedially near each side, with but few small irregular callous spots, the median line uneven, impunc- tate, feebly striate medially toward base but without distinct sub- basal puncture; elytra three-fourths longer than wide, two- fifths wider than the prothorax, the sides feebly converging and broadly sinuate, becoming somewhat gradually rounded and oblique, then nearly straight to the rounded apices, the sutural angle dentate; surface nearly as in the preceding but with much less unequal intervals either in width or in their generally feeble convexity; under surface less punctate, the pol- ished abdomen finely and sparsely so, the spots orange, those of the 98 CASEY inner series large and transverse, the external small but uniting with the inner on the subapical segments, those of the last segment very large, extending nearly to the apex but not transversely confluent. Length ($) 16.8 mm.; width 6.8 mm. Montana adonea n. sp. Prothorax at base almost as wide as the elytra, the sides almost a prolonga- tion of those of the latter, which are similarly rounded or obtuse at tip, with the sutural angle dentiform; surface shining, greenish to vio- laceous metallic, the anterior thoracic angles pale as usual 19 19 — Abdomen polished, very finely and sparsely punctate. Body nearly as in adonea; head ( 9 ) sparsely, irregularly punctate, solidly pale between the entire inner margins of the eyes, excepting the small anten- nal sinuses and two large and two minute discal central spots, the trans- verse hind margin of the pale area with a quadrate median emargina- tion; eyes moderate; prothorax evenly trapezoidal, with the sides nearly straight or very faintly biarcuate, the apex rather prominently rounded medially; surface broadly tumid sublaterally except near the apex, coarsely, rather closely and irregularly punctate, with but small and badly defined smooth spots, the median hne impunctate but not tumid, unipunctate basally ; elytra three-fourths longer than wide, nearly as in adonea but with more evidently unequal intervals; orange spots of the inner and outer abdominal series moderate, generally transversely united, those of the last segment transverse but not meeting medially and each prolonged posteriorly along the margin for a short distance. Length (?) r 7.0 mm. ; width 6.9 mm. Utah (Stockton) . . .histrion. sp. Abdomen more closely, more or less strongly and conspicuously punctate though somewhat variably so as usual 20 20 — Form rather broad, only moderately convex, shining, dull metallic green to violaceous; head (9) strongly but not very densely punctate, the back part of the front between the upper part of the eyes less punc- tate and medially feebly impressed, the front surrounded broadly with orange-yellow, which is sometimes solid excepting two median spots; eyes feebly convex, not prominent; prothorax trapezoidal, two-thirds wider than long, the sides broadly, very feebly biarcuate; surface almost evenly convex, rather coarsely but unevenly punctate, with the usual im- punctate areas sublaterally and a smooth median line, which is frequently striatulate along the middle posteriorly, the fovea subobsolete; elytra evidently less than twice as long as wide, parallel, rather rapidly round- ing and oblique at the sides in apical two-fifths, the apices rounded, becoming transverse suturally, the sutural angle dentiform; surface with the intervals strongly alternating in convexity, sparsely and very inconspicuously punctate throughout, the subhumeral marginal cavity deep; under surface strongly, moderately coarsely and closely punctate, immaculate except the abdomen, where the four series of yellow spots are well developed, the two lateral transversely confluent subapically, the two large spots of the last segment not attaining each other medially. Length (9) 18.0-19.5 mm.; width 6.8-7.4 mm. New Mexico (Fort Wingate), Colorado (Salida), Arizona (Grand Canyon of the Colorado) and Idaho (Coeurd'Alene — the color brighter aeneous) . .subornataLec STUDIES IN THE AMERICAN BUPRESTID.E 99 A — Nearly as in subornata but more elongate, more convex, with less unequally convex elytral intervals and with the sides of the elytra much more gradually rounding posteriorly and thence more gradually oblique and more evidently arcuate to the apices, the latter broadly rounding, very oblique thence to the strongly dentate sutural angles; head and prothorax nearly similar, except that the latter is less trans- verse, the elytra more elongate; abdomen similarly deeply sulcate along the median line of the first segment, coarsely, densely and very conspicuously punctate throughout, the spots as in subornata but deeper orange-red in color. Length (9) 18.5 mm.; width 6.9 mm. Colorado (Boulder Co.) punctiventris n. subsp. Form narrower than in subornata but only slightly more convex, shining, obscurely violaceous ; head ( 9 ) strongly but rather sparsely, irregularly punctate, similarly ornamented, the eyes rather more convex; pro- thorax nearly as in subornata, the punctures similarly sparse and irreg- ular but smaller medially, the sides broadly arcuate in basal half, thence nearly straight or less arcuate to the apex, the punctures becom- ing relatively coarser and denser near the sides than in subornata; scutellum oblong, parallel, flat; elytra nearly twice as long as wide, almost as in subornata, except that the surface is more punctured, the punctures less fine; abdomen distinctly punctured but much less coarsely or closely than in the two preceding forms, the spots small, red, those of the lateral pairs not approaching each other transversely, the two of the last segment very different, being transverse, not at all ramate posteriorly along the margin and uniting on the median line to form a single transverse fascia. Length (9) 17.2 mm.; width 6.5 mm. Kansas violescens n. sp. 21 — Form nearly as in the preceding, elongate, fusoid and notably convex; elytra strongly longitudinally undulated medially 22 Form more parallel and notably less convex; integuments black, never with the marked metallic lustre of the preceding section, the elytra more even and less undulated 23 22 — Body black-bronzed, punctured, glossy; head with a levigated eleva- tion in the centre between the eyes, channeled behind, marked with two yellow spots beneath on the cheeks, a triangular one adjacent to the eyes on the inner side and two dots of the same color between them; prothorax coarsely punctate, with smooth spaces, the discal one longitudinal, with an ante-scutellar impression; sides con- verging at the base of the prothorax; anterior angles with a yellow stripe; elytra strongly furrowed, with numerous very minute punctures in the furrows, rounded at apex; interstices with scattered punctures, those of strias 2-3, 4-5 and 8-9 elevated so as to form an obtuse ridge; four last ventral segments of abdomen having on each side a reddish-yellow spot, those on the last segment being the largest and most irregular; anus with two distinct lateral teeth. Length 18.5 mm. British Amer- ica (Latitude 54°, — Kirby , and one specimen vdthout more exact locality than"B. A." received from Chas. W. Leng); Washington State. rusticomm Kirby lOO CASEY Body nearly as in rusticorum, of the same color, the head, anterior thoracic angles and abdomen similarly spotted, except that on the first there are no frontal dots; the prothorax however is differently shaped, being somewhat constricted anteriorly, with the sides toward base rounded; it is also channeled [referring without much doubt to the impressed stria, frequently observable toward base in many species]; the sculp- ture of the elytra is similar but they are more attenuated and truncated at the end, as in rustica, and armed with three minute denticles. Length 15.5 mm. British America (Cumberland House, Lat. 54°). paganorum Kirby Body much smaller and narrower than in rusticorum, polished, brighter bronzed, greenish or subcupreous; head moderate, densely punctate, with median and subposterior levigated spaces; eyes only moderately convex; prothorax three-fourths wider than long to nearly twice as wide as long, trapezoidal, the sides more or less inflated and broadly arcuate in rather more than basal half, the surface almost evenly convex, strongly but very unevenly punctate, coarsely and densely so laterally, more sparsely and less coarsely medially, with large sublateral smooth spaces and a distinct smooth median line which is generally finely, briefly striate at base; elytra but little wider than the prothorax, nearly twice as long as wide, gradually strongly oblique at the sides in apical third, the apices transversely truncate, the truncature sometimes irregular or finely spiculose, the outer angle obtuse, never dentiform; under surface strongly, closely punctate; lateral spots of the penulti- mate ventral segments very small and frequently subobsolete, those of the last segment obliquely linear, sometimes also posteriorly ramate along the margins but never meeting on the median line. Length (c?, 9) 14. 0-16.0 mm.; width 5.0-5.9 mm. New Hampshire, Ver- mont, Michigan, Wisconsin (Bayfield), South Dakota and Mani- toba. [ = sexnotata Lap.-Gory] maculativentris Say 23 — Pale spots well developed at the sides of all the ventral segments as in the preceding group 24 Pale spots wanting or greatly reduced at the sides of the abdomen, except on the last segment, where they are correspondingly more conspicuous, forming a large transverse red fascia 27 24 — Prothorax very strongly narrowed anteriorly, the apex not more than two-thirds as wide as the maximum width. Body elongate, convex, black, the pronotum with very feeble cupreo-violaceous lustre; head ( 9 ) densely punctate, the intercular surface in great part coarsely rugose, with the entire frontal margin, enclosing two black points and emitting a long posterior medial spur, pale yellow; also with a suboblique and more orange spot at the upper part of each eye, the eyes not prominent; antennae rather long, extending fully to the tho- racic base; prothorax nearly twice as wide as long, widest behind the middle, the sides evenly and strongly arcuate, becoming gradually very convergent and less arcuate anteriorly; surface with the usual antero-lateral yellow vitta, almost evenly convex, coarsely, densely STUDIES IN THE AMERICAN BUPRESTIDiE lOI but unevenly punctate, with sublateral smooth spaces and a narrow and uneven impunctate median line; elytra long, but little wider than the thoracic disk, twice as long as wide, the subparallel sides very gradually rounding and oblique posteriorly, the aj)ices rounded, with the sutural angle scarcely at all prominent; surface with the intervals alternatingly feebly and strongly convex, moderately punctate, rather coarsely and closely so and subrugose laterally; abdomen deeply and abruptly sulcate basally, rather strongly but sparsely punctate, the yellow spots distinct and subequal on every segment, larger and trans- verse though separated on the fifth. Length (9) 19.5 mm.; width 7.0 mm. New Mexico (Cloudcroft) acomana n. sp, Prothorax not so strongly narrowed and less rounded on the sides, the apex three-fourths to four-fifths as wide as the maximum width; elytral apices similar 25 25 — Elytra oblique at the sides in fully apical two-fifths, more prolonged apically, the eyes rather more prominent. Body oblong, rather de- pressed, deep black; head (9) coarsely rugose, with a large median area somewhat depressed and with a short callous median line, finely and densely punctate, the front wdth four anterior pale spots; epistoma medially unusually deeply sinuate; antennae rather thick; prothorax short, almost twice as wide as long, the sides converging, evenly and moderately arcuate from just before the base to the apex, rounding in slightly at base; apex and base broadly bisinuate; surface coarsely and irregularly punctured as usual, the smooth median line with a striaform impression at base; scutellum longer than wide, metallic; elytra not quite twice as long as wide, scarcely at all wider than the prothorax, with the usual nearly flat, alternating with convex, intervals, the flat ones much the broader; punctures fine and sparse; abdomen rather strongly and closely punctate but not coarsely, the basal sulcus moderate, not quite attaining the apex of the first ventral; spots red, those of the fifth segment transversely subquadrate, widely separated. Length (9) 18.0 mm.; width 7.0 mm. Kansas morosa n. sp Elytra generally oblique in about apical third; eyes not prominent 26 26 — Form rather more elongate and scarcely so depressed, black, sometimes with very faint greenish or violaceous lustre either on the elytra or pronotum, the former frequently dark piceous; head nearly as in morosa but with more yellow maculation, this sometimes solidly surrounding the median parts of the front though broadly open behind, the epistoma less deeply sinuate, the antennse rather longer; prothorax much less transverse, three-fourths wider than long, the sides frequently a little more rounding basally, the sculpture coarse, nearly similar; elytra nearly similar, though a little more elongate and more evidently wider than the prothorax, the punctures larger, closer and much more con- spicuous especially laterally; abdomen as in morosa, the pale spots of the last segment rarely coalescent transversely; male apparently very rare. Length (9) 17.0-20.0 mm.; width 6.6-7.8 mm. Col- orado,—Cab. Levette fusca n. sp. I02 CASEY A — Nearly as in fusca but shorter and more oblong, the head smaller but otherwise nearly similar, the eyes not at all prominent, the cen- tral punctuation less fine and rather less crowded, the antennae somewhat thicker; prothorax more transverse, not quite twice as wide as long, the sides broadly, subevenly rounded, more converging but still arcuate anteriorly,the punctures rather sparser ; scutellum rounded ; elytra shorter, evidently less than twice as long as wide and not very evidently wider than the prothorax, otherwise similar; abdomen as in fusca but somewhat less strongly and decidedly sparsely punctate, the lateral yellow spots smaller throughout. Length (?) 17.0 mm.; width 6.7 mm. Montana sublivida n. subsp. Form nearly as m fusca but much smaller in size, shining black, the pronotum sometimes violaceous; head nearly similar but with less evident and much smaller central punctate area, the front yellow (c^), excepting two large posterior areas, or with the yellow area more contracted and broken into spots (9); eyes sUghtly convex; antennas rather long; prothorax nearly twice as wide as long to distinctly less; the sides sometimes slightly prominent near basal third but generally almost even; surface less coarsely and more densely punctate than in morosa though smilarly irregular, slightly impressed sublaterally behind the middle; scutellum rounded; elytra not quite twice as long as wide, but little wider than the prothorax as a rule, sculptured as m fusca, with almost equally distinct punctuation though rather less strongly alter- nating intervals; abdomen shining, alutaceous, strongly, rather closely punctate, the orange lateral spots distinct, subequal, those of the fifth segment somewhat obHquely transverse, moderate, basal sulcus strong, the second segment also sometimes slightly impressed medially. Length (c?, 9) 14.0-17.0 mm.; width 5.2-6.8 mm. Colorado (Boul- der Co. and Ouray) and New Mexico (Albuquerque and Cloudcroft) . caliginosa n. sp. 27 — Pronotum with four rounded impressions near the base, one on each side of the basal sublateral callus. Head rather small, densely punc- tate and somewhat impressed medially on the front, with several small smooth patches and with well developed yellow maculation, the eyes not prominent; antennae rather slender; prothorax small, two-thirds wider than long, widest at basal third, where the sides are prominently rounded, thence rather strongly converging anteriorly and less strongly posteriorly and nearly straight; apex and base bisinuate; surface coarsely punctate, densely laterally, the sublateral smooth spots prominent, the smooth median line irregular, feebly striate basally; elytra almost a third wider than the ])rothorax, less than twice as long as wide, parallel, oblique but rounded at the sides in apical third, the apices truncate but with obtuse angles; surface with the equal intervals slightly alternating in convexity, the punctures sparse as usual but strong and distinct; abdo- men strongly, rather closely punctate, narrowly sulcate at base; seg- ments two to four each with a small lateral spot, the fifth with a large entire fascia which is sinuate at the middle of its anterior margin, arcuato-truncate at tip, with rounded angles. Length (9) 16.7 mm.; width 6.5 mm. Montana nigricans n. sp. STUDIES IN THE AMERICAN BUPRESTID^ IO3 Pronotum with a deep foveiform impression at each side of the median line before the middle but otherwise not distinctly impressed. Surface shining, greenish-black; head (c?) coarsely, densely punctate and par- tially impunctate, the entire space between the eyes yellow-orange, exxept a large median anteriorly and deeply incised black spot; eyes large but not prominent; antennas slender; prothorax short, almost twice as wide as long, only moderately narrowed anteriorly, the sides subevenly rounded, a little more so basally, widest at about basal third; surface strongly, sparsely punctate, densely so laterally, the sub- lateral smooth spots coalescent, not prominent, with traces of the subbasal depressions of the preceding; anterior sides and lateral third of the apical margin pale; elytra nearly as in nigricans but only very slightly wider than the prothorax, the apices subtruncate and with feeble minute irregularities but with the outer angle obtusely rounded ; surface strongly but not closely punctate, the intervals alternatingly rather wide, moderately convex, less polished and feebly rugose and narrower, strongly convex, smooth, less punctate and very lustrous, the arrangement very regular throughout the width; abdomen deeply sulcate basally, strongly, rather closely punctate, the third and fourth segments each with a very small red dot laterally, the fifth with two large transverse red spots, the apex evenly and distinctly sinuate with simple angles. Length (d^) 18.5 mm.; width 6.9 mm. British Columbia (Golden) lyrata n. sp. Pronotum without evident impressions. Body large and elongate, shining, black or greenish-black to faintly violaceous in lustre; head nearly as in lyrata, sometimes vnth a larger finely and densely punctate medial area, the pale area surrounding the interocular surface nearly similar and frequently as well developed in the female as in the male; prothorax larger, much less transverse, generally but little narrower at base than at basal third, the sides more even thence to the apex, nearly straight or feebly arcuate, sometimes feebly biarcuate, but, in the male, some- times subangularly prominent behind basal third; surface more densely punctate, with the sublateral smooth areas usually large though not prominent; elytra but little wider than the prothorax, generally fully twice as long as wide, the apices truncate to arcuato-truncate, nearly even, the sutural angle unusually evident for this section; surface nearly as in lyrata but less strongly punctate, the wider intervals more nearly flat and not so rugulose, the narrower less convex; abdomen strongly but not very coarsely, closely punctate, sulcate at base, the fifth seg- ment with a large and generally entire transverse red fascia, the two or three preceding also sometimes with a small red dot at each side, the fifth broadly sinuate at tip (c^), or arcuato-truncate with obtuse angles ( 9); females abundant, the male rare. Length (c?) 15.0, ( 9) 18.0-23.0 mm.; width (c?) 5.8, (9) 7.0-8.8 mm.; dimensions of one 9 23.0 by 8.3 mm., of another 9 22.0 by 8.8 mm.. Oregon, Idaho (Coeur d'Alene), Washington State and Vancouver Lsland). [=rusti- corum Lee. nee Kirby]. adducta n. sp. 28 — Form moderately elongate and subdepressed, vivid green above, the I04 CASEY elytra violaceous blue to green, varying with angle of light, thickly and uniformly, finely spotted with fulvous yellow, the spots more or less transversely confluent, a transverse band behind apical third, which is continued broadly on the suture to the apex, devoid of spots; under surface ( $ ) coppery with greenish reflection, without paler maculation, or (c?) having the median parts from the head to near the apex of the first ventral pale yellow, with a small yellow spot at each side of the fifth ventral; head (9) without pale spots, rather coarsely, densely punctate, the eyes rather prominent, or (c^) ver)' densely and more finely punctate; prothorax two-thirds wider than long, the sides evenly converging, almost evenly and moderately arcuate from base to apex; surface even, evenly convex, without impunctate areas, the smooth median line interrupted anteriorly, the punctures coarse, very dense laterally; elytra but little wider than the prothorax, parallel, nearly twice as long as wide, the sides gradually oblique in more than apical third, the apices obliquely rounded, the sutural angle not at all promi- nent; surface with equal, moderately convex, finely and sparsely punc- tate interv^als, the stride closely and evidently punctate; abdomen very finely, sparsely punctate, the first segment narrowly flattened along the median line. Length (9) 16.0 mm.; width 5.7 mm. Kansas (Ed- mond) , — Warren Knaus. [ = confluens Lee] confluenta Say A — Nearly as in confluenta throughout, the arcuate elytral band free from maculae not prolonged posteriorly on the suture, the prothorax shorter, with the lateral punctures still more confluent, the impunctate me- dian fine entire, the side margins with a pale point at the apical angles and another near the base; head (d^) finely, very densely punctate and much more pubescent; last ventral ((^) with an entire transverse pale fascia from side to side near the base, the punctures stronger than in the female of confluenta and less sparse. Length (c?) 14.7 mm.; width 5.0 mm. Texas tessellata n. subsp. 29 — Elytral striae more or less fine, with small, even, less close-set punctures. Atlantic regions, extending westward nearly to the Rocky Mts 30 Elytral striae sulciform, densely and more confusedly crenate, the inter- vals relatively narrow; median line of the pronotum generally more impressed. Pacific regions, extending eastward in the north as far as Lake Superior 34 30 — Body very elongate, in great part pale beneath, the legs at least in part pale, the elytra each with an anterior vitta from the basal margin 31 Body less elongate, without paler areas beneath, the legs always black with more or less metallic lustre, the elytra variously maculate with yellow. .32 31 — Legs entirely pale in color. Body only moderately convex, shining, metallic green, the anterior parts frequently bluish to aeneous; lateral parts of the sterna metallic green, the remainder of the under surface variegated with yellow and brownish-red, the latter prevailing on the abdomen, the former forming a broad but broken axial vitta from the head to the middle of the abdomen, apparently in both sexes, the sides STUDIES IN THE AMERICAN BUPRESTID^E I05 of the body beneath also with yellow spots, forming a broad fascia on the last ventral; head moderate, the eyes rather prominent, the front without pale spots, coarsely, closely punctate, with a prominent smooth callus along the median line centrally; antennae slender, pale brown; prothorax two-thirds wider than long, but Httle narrower at apex than at base, the former truncate, the latter bisinuate; sides evenly, feebly arcuate; surface evenly convex, coarsely, not densely punctate, without smooth spots or median line but with a small fovea before the scutellum, the hypomera pale, the latter tint sometimes spreading more or less broadly to the dorsal surface; elytra about two and one-half times as long as wide, only just visibly wider than the prothorax, very gradually rounded and narrowed posteriorly behind the middle, the apices rather broadly sinuate and strongly bispiculose, each with a yellow vitta from just beyond the middle of the base for two-fifths or more, a transverse discal fascia at apical two-fifths and another still more posterior, the apices also pale; stria; impressed, punctulate, the intervals equal, moderately convex, becoming flat bas- ally, finely, sparsely punctate; under surface finely, sparsely punctate, more strongly and closely laterally. Length (c?, $) 17. 5-21.0 mm.; width 5.4-6.7 mm. Pennsylvania, Indiana, Kentucky, Louisiana and Kansas ruflpes Fabr. A — Similar to rufipes but much larger and stouter, the head more devel- oped, with the front much more convex, the eyes fully as large but not prominent, their contour following the general outline; scutellum equally well developed, flat, rounded; elytra with the sides much more rounding and converging posteriorly, the sinuate apices rela- tively narrower though strongly bispiculose. Length 23.0 mm.; width 7.4 mm. One specimen labeled "Cal." in Cab. Levette, but probably in error virens n. subsp. Legs black and metallic bluish-green, the femora pale, vnth the upper side, a lateral streak and the tip metallic black, the anterior pale only on the under surface in basal two-thirds. Body rather less slender, shin- ing, black to greenish-black, the under surface in great part pale; head coarsely, densely punctate, with a few callous smooth spots, a large, posteriorly cuneate median pale spot at apex and another at the lower part of each eye, the latter not at all prominent ; antennae black with feeble metallic lustre; prothorax as in rufipes, the sides broadly pale, with a dark vitta on the hypomera; scutellum very much smaller; elytra as in rufipes but less elongate and evidently wider than the pro- thorax, very gradually and arcuately narrowed posteriorly to the apices, which are briefly produced and each rather wide, deeply, obliquely sinuate, the external angles strongly spiculate, the sutural much pro- duced and acute, similarly striate, the intervals not distinctly punctate, each elytron with a narrow reddish vitta from the humeral angles over the callus to two-fifths from the base at outer third, a small discal spot just behind the middle, another larger, more transverse and lateral at apical fourth and a small longitudinal dash near the suture between the last and the apices, the latter not pale; under surface completely I06 CASEY impunctate, except some very fine sparse asperate punctules laterally, in great part pale, each abdominal segment with a very large irregular pale fascia which is sometimes medially interrupted, the fifth wholly pale, except the tip and the lateral basal angles. Length 20.0 mm.; width 6.5 mm. One specimen from Cab. Levette labeled "N. Y.," but more probably from Colorado elongata n. sp. 32 — Elytra each with three large discal pale spots, the anterior elongate, oblong or oval and not attaining the base, the second transverse, at three-fifths, extending nearly from suture to side margin, the third feebly oblique, discal and near the apex. Body subparallel, rather depressed, brassy-green, the elytra somewhat more obscure, the under surface green with cupreous reflection; head strongly, densely punc- tured, without callous spots even medially; antennce black, but feebly metallic; prothorax three-fourths wider than long, the apex but slightly narrower than the base, the sides subevenly, feebly arcuate; surface without pale maculation of any kind above or beneath, coarsely and not very closely punctate, without smooth or impressed line, with a small ante-scutellar puncture; scutellum small, subtransverse, oval; elytra twice as long as wide, distinctly wider than the prothorax, par- allel, arcuately narrowed behind, the apices sinuate and strongly, acutely bidentate; surface finely striate, the striae as mfasciata, punc- tate, the punctures more visible viewed obliquely from the side than from the suture, the intervals nearly flat, equal, finely and remotely punctate; abdomen finely punctate, densely so laterally. Length 11.3 mm. ; width 4.0 mm. Michigan sexplagiata Lee . Elytra each with a conspicuous irregular pale fascia at three-fifths and a small oblique subapical spot, also sometimes a small discal pale spot at two-fifths from the base 2;^ 33 — Sides of the prothorax incurvate at base and feebly swollen basally, but thence moderately converging and nearly straight to the apex. Body oblong-elongate, moderately convex, polished, cupreous, generally more greenish anteriorly, sometimes obscure throughout, the under surface bright cupreous to greenish or even violaceous in the dark forms; head strongly but very unevenly punctate, the eyes prominent; prothorax nearly twice as wide as long, the apex about four-fifths as wide as the base, broadly, subevenly sinuate, the base feebly bisinuate; surface coarsely, sparsely punctate, more closely laterally, feebly to not at all impressed along the middle, generally with a small sublateral rounded impression, the ante-scutcllar fovea conspicuous; scutellum moderate, variable; elytra slightly more than twice as long as wide, just visibly wider than the prothorax, parallel, arcuately narrowed behind, the apices sinuate and strongly bispiculose; striiE rather strong, evenly and closely punctate, the punctures on the inner slope of the grooves, the intervals nearly flat though slightly incHned sutureward, extremely minutely, remotely punctulate; abdomen with the basal segment very faintly and indefinitely impressed along the middle, finely, not very closely punctate, more strongly, densely so laterally; basal joint of the hind tarsi unusually short, verj' much shorter than the STUDIES IN THE AMERICAN BUPRESTID^E IO7 next two combined. Length 14.8-15.7 mm.; width 5.2-6.0 mm. Dakota, Wisconsin (Bayfield) and Indiana fulgenH n. sp. Sides of the prothorax evenly converging and distinctly arcuate from a short and very feeble though abrupt basal swelling to the apex, not evidently incurvate at base; subapical spot of the elytra very much more oblique. Body more elongate in form, brilliant and pure green in color, becom- ing subajneous beneath, maculate nearly as in fulgens but with the jjost- median fascia generally fuller and less irregular; head strongly and closely but somewhat irregularly punctate, the eyes moderately promi- nent; antennae piceous-black, with feeble metallic reflection; prothorax almost twice as wide as long to evidently less, more or less feebly im- pressed along the median line, strongly and subequally but not very closely punctate throughout, with a small and feeble sublateral impres- sion at basal third; scutellum rather small, subquadrate but variable as usual; elytra slightly more than twice as long as wide, evidently wider than the prothorax, parallel, the sides gradually rounding and con- verging in apical third, the apices, stria) and punctuation nearly as in fulgens; a small ante-median spot sometimes present but frequently wholly obsolete; all the spots, as in fulgens, surrounded by a black clouded border; abdomen punctulate nearly as in fulgens, the lateral punctures similarly larger and rather dense, only slightly more asperu- late. Length 14. 2-1 7.5 mm.; width 5.2-6.4 mm. Pennsylvania, New York and Ontario (Sudbury). [= sexmaculala Hbst., and /Aerwin- ieri Chev fasciata Fabr. A — Similar but more elongate, bright cupreo-a,'neous above and below, with feeble greenish reflection at the sides of the elytra, the latter similarly maculate and sculptured; head nearly similarly sculptured; prothorax less transverse, three-fourths wider than long, equally strongly but much more densely punctured throughout, feebly im- pressed medially toward base; elytra nearly two and one-half times as long as wide the sides more gradually rounding and converging behind and less oblique than in fasciala, the apices similar; abdomen similarly punctulate, the first segment more strongly impressed along the middle though indefinitely as in fasciala. Length 16.2 mm.; width 5.7 mm. Michigan fortunata n. subsp. 34 — Elytra each with three -large and distinct pale spots, one near basal third and sometimes extending to the base, another behind the middle and the third near the apex, the apices always sinuate and acutely bidentate, the pronotum more or less evidently impressed along the middle 35 Elytra immaculate or having sporadically a small spot near three-fifths and another still smaller near the apex 39 35 — Anterior pale spot extending obliquely to the base and enveloping the humeral callus. Body elongate, brassy-green; head purplish, punctate, the front carinate; prothorax one-half wider than long, punctured, the sides subparallel, the base bisinuate; elytra purplish, the basal pale spot including a dark spot on the humeral callus, the spot at about one- I08 CASEY third from the apex transverse, not attaining the suture but extending to the margin, the subapical extending nearly to the suture and dilated along the margin, the spots reddish-yellow in color; striae deep, punc- tate, the intervals sparsely punctate; antennae testaceous, the first joint asnescent. Length 15.0 mm. Washington State (Steilacoom) . gibbsi Lee. Anterior pale spot wholly discal like the others, never extending to the base; antennas always black, with more or less aeneous or greenish metallic lustre throughout 36 36 — Subapical spot of the elytra very oblique, the anterior spot nearer the side margin than the suture. Body brilliant and pure green, becoming faintly aeneous medially beneath, slender; head very densely, sub- uniformly punctate; eyes prominent; prothorax four- fifths wider than long, the sides parallel and straight to beyond the middle, there rounding and distinctly convergent to the apex, the latter broadly sinuate, feebly arcuate medially; surface bluish centrally, coarsely, closely punctate, confluently so laterally; elytra nearly two and one-half times as long as wide, a fifth wider than the prothorax, parallel, the sides rounding and becoming oblique in apical third, the striae deep, coarse, finely, con- fusedly punctate, the intervals not quite twice as wide as the striae, remotely punctulate, not very convex; anterior spot moderate, rounded, almost twice as far from the suture as from the side, the second trans- verse, irregular, not attaining either suture or sides, the third not attaining the sides or suture; abdomen with the basal segment not at all impressed, finely punctate, densely so laterally ; legs slender. Length 13.5 mm.; width 4.5 mm. California bistrinotata n. sp. Subapical pale spot only moderately obUque, the anterior on the median line 37 37 — Prothorax apparently parallel, the sides evenly and distinctly arcuate. Body very slender, slightly convex, colored as in bistrinotata; head small, densely punctate, with a small central aeneous callus not sug- gested in the preceding; eyes prominent; prothorax short, very nearly twice as wide as long, strongly but sparsely punctured, more closely in the medial depressed line and more coarsely as well as densely later- ally, with a deep rounded sublateral impression behind basal third, the apex but little narrower than the base; scutellum rounded, convex, smooth; elytra fully two and one-half times as long as wide, between five and six times as long as the prothorax and nearly a third wider, nearly as in the preceding but more broadly sinuate at the tips; inter- vals narrower and more convex, the spots all discal, not attaining suture or sides; abdomen finely, more sparsely punctate, the basal seg- ment very faintly impressed along the middle, the fifth segment trun- cate and feebly bidenticulate at apex in the male. Length 12.5 mm.; width 4. 1 mm. California angusta n. sp. Prothorax similarly distinctly narrower than the elytra but with the sides converging from base to apex, broadly arcuate basally, becoming straighter anteriorly or evenly arcuate throughout 38 STUDIES IN THE AMERICAN BUPRESTID^ IO9 38 — Sides of the prothorax evenly arcuate from base to apex. Body very elongate, moderately convex, shining, viridescent, with very strong cupreous lustre, the under surface brilliant cupreous-red medially and throughout the abdomen, viridescent laterally and with unusually long ashy pubescence; head very densely and confluently but subuni- formly punctate, the eyes large and prominent; prothorax three-fifths wider than long, the apex decidedly narrower than the base, obsoletely bisinuately truncate; surface strongly, closely punctate, densely and more coarsely so laterally, with the impression large, shallow and dif- fuse, the median line broadly but strongly impressed throughout; elytra between four and five times as long as the prothorax and a fourth wider, parallel, the sides rounding and converging behind in apical third, the apices sinuate and acutely bidentate; sculpture as in histri- notata but with the punctures of the inner of the two confused series of the striae more distinct; anterior pale luteous spot large, longer than vAde and anteriorly pointed, the second large, almost rounded, the third broad, attaining neither suture nor sides; abdomen with the basal segment feebly flattened medially, the punctures fine but strong, dense laterally, the fifth segment broadly, angularly emarginate at tip in the male. Length 16.0 mm.; width 5.2 mm. Oregon (New- port),— H. F. Wickham callida n. sp. Sides of the prothorax gradually more arcuate basally. Body less elongate, parallel, subdepressed, shining, green, sometimes with very evident cupreous lustre, cupreous beneath on the abdomen, otherwise green laterally; head densely, confluently punctate, the eyes prominent, pale in examples at hand; prothorax shorter, three-fourths wider than long, the apex much narrower than the base ; surface broadly impressed medially, strongly, rather closely punctate, subconfluently so laterally, without evident impression; base with a small truncate and abruptly formed lobe at the scutellum, which is not at all evident in callida, the ante-scutellar puncture distinct; scutellum small, rounded, variable; elytra as in callida though shorter, but little more than twice as long as wide, the sides more arcuate to the tips, which are sinuate and bispicu- lose as usual; sculpture and maculation nearly similar, as is also the punctuation of the under surface, the anterior spot much smaller, the post-median more transverse; apex of the fifth ventral transversely truncate in the male. Length 14. 5-15. 5 mm.; width 5.0-5.6 mm. California fastidiosa n. sp. A — General characters relating to form, coloration, maculation and sculpture as in fastidiosa, the eyes still more prominent; prothorax notably shorter and more evidently narrower than the elytra, almost twice as wide as long and rather more sinuate at apex, but otherwise similar; elytra even more rounded on the converging sides in apical third, the sinus of the apices very much narrower than in any other form of this group; abdomen much less cupreous and only faintly so medially, broadly green laterally. Length 13.0 mm., width 4.6 mm. California mediocris n. subsp. 39 — Elytral apices more broadly and frequently obliquely sinuato-truncate I lO CASEY at tip, with the outer angle strongly dentiform, the sutural acutely prom- inent 40 Elytral apices much narrower, feebly sinuato-truncate, with the outer angle obtuse, sometimes almost rounded, the sutural acute though seldom very noticeably prominent 42 40 — Prothorax but little narrower than the elytra. Body elongate, slightly convex, shining, green, sometimes vnth decided cupreolate reflection, except the abdomen and median parts of the under surface anteriorly, which are brightly polished and cupreous; head rather small, densely, subconfluently punctate and somewhat rugulose, the eyes not very prominent; antennae black with metallic lustre; prothorax two-thirds or more wider than long, the sides strongly converging anteriorly, more rounded basally; surface strongly and unevenly but rather sparsely punctate, confluently so in places laterally, broadly and very feebly impressed along the middle, the ante-scutellar puncture very small or obsolete; elytra rather evidently more than twice as long as wide, par- allel, the sides gradually rounding in apical third to the apices, coarsely furrowed, the furrows confusedly crenate; intervals narrow, remotely punctulate, generally wholly immaculate but occasionally having each a small spot near apical third and a minute subapical dot; abdomen very minutely, sparsely punctate, less finely, closely so laterally. Length 16.0-19.2 mm.; vndth 5.2-6.3 mm. California and Oregon, — abun- dant crenata n. sp. (Mots., i. litt.?) A — Form and coloration throughout as in crenata, the head similar but relatively a little larger; prothorax differing very obviously, being smaller and less convex, shorter and more transverse, the sides less inflated basally, the central portions bluish and the sublateral rounded impressions notably more posterior in position; elytra nearly similar throughout but more distinctly wider than the prothorax; under sur- face similarly minutely and remotely punctulate, with broadly convex prosternum and transversely truncate fifth ventral. Length 16.5 mm.; width 5.9 mm. California (Yreka) seditiosa n. subsp. Prothorax much narrower than the elytra 41 41 — Size large, notably depressed, viridi-cupreous, in great part bright cu- preous beneath, the external marginal line of the elytra finely cupreous; head small, strongly, coalescently punctato-rugose, the eyes large but not prominent; antennie black, metallic basally; prothorax two-thirds wider than long, the sides arcuate, converging anteriorly, the apex much narrower than the base, broadly sinuate; disk a little wider just behind the middle than at base, broadly and feebly impressed medially except toward apex, the subbasal foveola very feeble, the sublateral rounded impression behind the middle strong; punctures deep and rather coarse, moderately close, coarser, and crowded laterally ; basal angles right ; elytra a little more than twice as long as wide, almost a fourth wider than the protliorax, sculptured as in the preceding, not maculate; under surface almost similar, the first segment feebly flattened along the middle; legs slender, the basal joint of the hind tarsi not quite as long as the STUDIES IN THE AMERICAN BUPRESTID^ III next two combined. Length ( $ ) 19.5 mm. ; width 6.8 mm. Washing- ton State (Tacoma) leviceps n. sp. Size much smaller, shining, almost pure vivid green above, the under sur- face, except laterally, and the abdomen ceneo-cupreolate; head relatively somewhat larger, strongly, confluently punctate, the eyes moderate, not noticeably prominent; antenna? black, with the metallic lustre greatly diminishing apically; prothorax nearly four- fifths wider than long, al- most parallel, the sides but feebly arcuate though more rounding and converging anteriorly, the apex less evidently narrower than the base; surface strongly punctate, densely laterally, with four small rounded impressions, one on each side of the broadly and feebly impressed median line just before the middle, the others in the usual sublateral position but feebler than in the preceding; elytra slightly more than twice as long as wide, the sides more rectilinearly oblique apically, five times as long as the prothorax and almost a third wider, sculptured as in the preceding, immaculate; under surface similar, the fifth ventral in the male rather narrowly truncate at tip, the truncature very faintly bisinuate. Length (cJ*) 15.3 mm.; width 5.3 mm. California, — Cab. Levette depressa n. sp. 42 — Body beneath coarsely punctured; elytra cupreous at the sides. Form elongate-oval, subdepressed; head moderately densely punctate, brilliant green in color changing to cupreous ; prothorax broader than long, nar- rower in front, the sides very feebly arcuate; surface moderately convex, with rather coarse punctures, which are more densely placed at the sides, brilliant cupreous in color ; elytra at base wider than the prothorax, oblong- oval, the apex feebly attenuate, the tips feebly emarginate, with the sutural angle acute and slightly prolonged; surface striate, the intervals equal, moderately convex and sparsely punctate, the coloration diversified, brilliant blue along the suture, shading to green, with the margin cu- preous ; under surface very densely punctured at the sides of the metaster- num and abdomen, brilliant cupreous in color with green reflection; prosternum and first ventral not sulcate. Length 15.0 mm. Cali- fornia (Owen's Valley) and Oregon connexa Horn Body beneath finely punctured, more closely and strongly laterally as usual; elytra sometimes suffusedly cupreolate but never with distinct cupreous side margins 43 43 — Species of the Pacific coast regions 44 Species of the far interior of the continent 45 44 — Prothorax but slightly narrowed anteriorly, the sides almost parallel; upper surface pure brilliant green, the under surface cupreolate as usual. Body depressed, very shining; head very densely and conflu- ently but almost evenly punctate, the eyes not prominent; prothorax fully three-fourths wider than long, the sides nearly straight and par- allel, gradually more rounding and converging anteriorly; surface coarsely punctate, confluently so laterally, feebly impressed on the median line, the sublateral rounded impression behind the middle Proc. Wash. Acad. Sci., April, 1909. 112 CASEY very deep and conspicuous; elytra between two and three times as long as wide, almost a third wider than the prothorax, straight and parallel at the sides, the latter gradually oblique and becoming straight in apical two-fifths, the sutural angle not prolonged though acute; surface deeply crenato-sulcate as usual, with a small pale spot only slightly behind the middle and another at apical sixth; under surface finely, sparsely punctate, rather closely and strongly so laterally. Length 15.0 mm.; width 5.2 mm. British Columbia (Stickine Riv.), — H. F. Wickham. viridimicans n. sp. Prothorax much narrower at apex than at base, the sides converging and almost evenly arcuate throughout; upper surface golden-green. Body oblong, depressed; head and disk of the prothorax ccerulescent, [the form of the latter assumed as above] ; head rugose, the pronotum deeply punctate; elytra densely costate, the interstices [grooves] rugulose; apex truncate, unarmed. Length 17.0 mm.; width 6.0 mm. Alaska (Sitka) langi Mann. A — Bright golden-green; head and prothorax thickly punctured, the latter with an impressed longitudinal line; elytra with deeply impressed punctured lines, having a purplish tinge on each side in front, full four times the length of the prothorax, cupreo-bivittate; abdomen with a short gilded band on the fore border of each segment, the tip cupreous. Length 18.0 mm. Vancouver Island . . ornata Walker B — Elongate, parallel, feebly convex, cuprascent above, the pronotum more brilliant cupreous, the sutural and basal regions of the elytra greener; under surface throughout highly burnished, deep and rich coppery- red, with faint greenish tinge near the sides anteriorly; head coarsely, contluently punctate, the eyes prominent, black; antenna^ cop- pery-red basally; prothorax not more than three-fifths wider than long, the sides rather more rounded basally than usual and more incurvate at base; surface broadly and only very faintly impressed along the middle, with the sublateral impressions wholly obsolete, the strong punctures rather sparse, coarser, very deep and dense laterally; elytra about two and one-half times as long as wide, nearly five times as long as the prothorax but only slightly wider, the sides gradually rounding and converging in apical third to the narrowly sinuato- truncate apices, the sutural angle very acute but not much prolonged; surface with very coarse, confusedly cribrato-punctate grooves, which are about as wide as the remotely jjunctulate intervals, im- maculate, suffusedly cupreolate; under surface finely, sparsely punc- tate, minutely so medially; femora unusually minutely, sparsely punc- tulate. Length 16.0 mm.; width 5.8 mm. Vancouver Island (Nanaimo), — H. F". Wickham incolumis n. subsp. C — Similar to the last but more elongate and rather more depressed, green» the elytra sufTusedly cupreolate, the under surface bright coppery-rcd» more broadly and evidently green toward the sides anteriorly; head very densely but less coarsely punctate, the eyes smaller and less j)romincnt; j)rothorax nearly similar, excej>t that the median line STUDIES IN THE AMERICAN BUPRESTID^ II3 is strongly impressed throughout; elytra similar though relatively broader, nearly a fourth wider than the prothorax and each with very small pale points near apical third and seventh or eighth; under surface with the tme punctures a little stronger. Length 17.0 mm.; width 6.0 mm. Oregon oregona n. subsp. D — Still more depressed, polished, green with cupreous lustre, the elytra cupreous, with the broad sutural regions green; under surface as in oregona; head and eyes as in that form, the prothorax also similar though a little shorter, with dense sublateral areas of smaller punc- tures than in oregona; scutellum a httle larger; elytra shorter though more than twice as long as wide, a little wider behind the middle than at base, a third wider than the prothorax, the sides more rapidly rounding and oblique posteriorly and becoming more rectilinear than in either of the preceding and in nearly apical two-fifths; sur- face less coarsely sulcate, the grooves more finely though confusedly punctate, the intervals wider, nearly flat, still more finely and re- motely punctate than in the two preceding, each with a small pale spot only a little behind the middle and another near apical eighth ; under surface somewhat less finely punctate, the femora more evidently. Length 16.0 mm; width 5.8 mm, British Columbia (Golden) obliqua n. subsp_ 45 — Sides of the prothorax evenly and unusually strongly rounded, becom- ing strongly convergent and gradually straighter anteriorly before about • the middle. Body above very deep and uniform bluish-green through- out, the under surface green to bluish, with faint aeneous reflection laterally on the abdomen only; head not coarsely, very densely and coalescently punctate, with the median line somewhat elevated but not smooth; eyes not prominent; antennae black; prothorax much shorter than in any other allied species, fully twice as wide as long, the apex much narrower than the base, broadly and almost evenly sinuate, the base sinuate only near each side; surface scarcely visibly impressed medially, the basal fovea distinct, the sublateral rounded and densely sculptured impression rather small; surface strongly punctured, closely but not very confluently so laterally; scutellum small, elongate; elytra long, two and one-half times as long as vidde, only very slightly wider than the prothorax, the sides almost rectilinearly oblique in more than apical third, the narrow apices very feebly sinuato-truncate; surface with the deep, densely and confusedly punctate sulci as wide as the inter- vals, sometimes with purplish-blue clouds and occasionally with small post-median pale spots as usual; under surface medially minutely and sparsely punctate. Length 15.5 mm.; width 5.2 mm. New Mexico. patruelis n. sp. Sides of the prothorax parallel and nearly straight, gradually rounding and moderately convergent anteriorly from near the middle 46 46 — Form elongate, subdepressed, uniform deep green though not very bright above, brighter and with some aeneous reflection beneath; head small, very densely, confluently punctate, the eyes prominent, 1 14 CASEY black in all examples at hand; prothorax short, very nearly twice as wide as long, feebly impressed along the middle, the subbasal fovea well developed, the sublateral impressions feeble ( 6^), or deep and conspicu- ous ( 9 ), also sometimes with two additional submedial feeble impres- sions in the former, which are obsolete in the latter, sex, where however a small obUque impression is visible near the base at lateral fourth ; punc- tures strong, unevenly dense laterally; elytra shorter than in patruelis though much more than twice as long as wide, much more distinctly wider than the prothorax, with a tendency to post-medial inflation, the sides behind obliquely and subrectiUnearly converging, the apices very feebly sinuate and with broadly rounded or obtuse external angles; surface nearly as in patruelis, generally immaculate but occasionally with small post- median spots, especially in the female; under surface similar. Length 13. 5-16.0 mm.; width 4.7-6.0 mm. Colorado (Boul- der Co.) graminea n. sp. Form less elongate, similarly depressed, deep and uniform more brilliant green above, cupreous and shining beneath, becoming greenish at the sides anteriorly; legs bright green; head finely, very densely, subuni- formly punctato-rugulose; eyes prominent, pale; prothorax as in gram- inea but rather more narrowed anteriorly, the apex deeply and evenly sinuate; surface narrowly, feebly impressed along the median line and with a strong rounded impression toward each side at basal third, strongly punctate, coarsely and very confluently so laterally; disk coerulescent medially, as also in a broad area along the elytral suture disappearing toward tip; elytra scarcely more than twice as long as wide, about a fourth wider than the prothorax, somewhat abruptly, strongly, subrectiUnearly narrowed in nearly apical two-fifths; surface with the sulci not quite so coarse but confusedly and closely punctate, the intervals much wider than the sulci, finely, remotely punctate, each elytron with a transverse sinuous pale spot and a small and very oblique subapical spot nearly as in fasciata, the spots surrounded by a black clouded border; punctures of the under surface stronger than in the preceding species but not conspicuous; basal segment of the abdomen very faintly impressed medially as usual in this section. Length 14.7 mm. ; width 5.5 mm. Lake Superior (Isle Royale) saturata n. sp. "Elytra like aurulenta; prothorax flattened, side margin thickened, disk with a broad dorsal stripe, and oblique space each side smooth; prothorax and under surface thinly clothed with long soft white hair. Califor- nia,— Coll. Mniszech. " villosa Lee. Some of the synonymy suggested by Crotch and since perpetuated in catalogues, is of so extraordinary a nature that it is difficult to ac- count for it as a deliberately formulated scientific opinion; perhaps the most completely untenable of these suggestions is that which unites such diverse forms as langi and sexplagiata with fasciata. The species described under the name connexa, by G. H. Horn, is only STUDIES IN THE AMERICAN BUPRESTIDiE II5 tentatively placed in the above table, for the remarks appended to the original description are so ambiguous that it is quite impossible to state anything definite concerning its relationships. The author at first places it in a group with gibbsi and conjluenta, which are alto- gether dissimilar species, and he finally states that it has elytra some- what as in adjecta, which is certainly not to be inferred from the lan- guage of the description; adjecta having no striae in the usual accepta- tion of that term. The villosa, of LeConte, is appended in the com- plete language of the original description. The species is really not described, for while the few remarks concerning the prothorax would seem to signify alliance with Buprestis proper, as here limited, the elytra are said to resemble those of aurulenta. The name aurulenta Linn., has been until recently applied to two very different species, so that it is impossible to determine whether LeConte had in mind ornata Fab., or his own lauta. The locality "California" appears furthermore to be very doubtful. The species near nmculipennis are difficult to separate and my results are not given with entire confidence, the only opinion that can be held with certainty being that a number of distinct species are generally confused in our collections. The original descriptions of Kirby's species are alone utilized for the characters given in the above diag- noses. Nuttalli, by the maculation of the coxae, may be allied more closely to consularis than to alternans, but the abdominal spots, ranged in a single series at each side, shows that is different from either. The inflation of the basal part of the prothorax in nuttalli is assumed, as no mention of such a character is made by Kirby. Of rusticorum I have two specimens; the outline is evenly fusiform and the sides of the trapezoidal prothorax are inwardly arcuate basally, which may account for the rather ambiguous language of the original diagnosis; the under surface is strongly and closely punctate and the elytral apices are obtusely subtruncate or apparently almost broadly rounded; the last ventral in these two females is truncate at apex, the sides of the truncature feebly subdenticulate, but this is a variable character. There can be little or no doubt that these specimens represent the true rusticorum and it differs considerably from those following it in the table, such as fusca and adducta, which have generally been assumed to represent the species. Paganorum seems to be very closely allied to rusticorum and may be merely a subspecies. I have assumed to Il6 CASEY represent consular is a species not very common in the eastern States; the description of Laporte de Castelnau and Gory is evidently taken from an unusually large female, but the language " elytres tres-epi- neuses ^ I'extremite" is not strictly true of any example of the con- sularis group coming under my observation. The drawing also shows that the elytra are more prolonged at tip than in any that I have seen. Elongata, of the above table, is evidently allied to the Mexican variegata, differing in the arrangement of the elytral spots, in its more obscure coloration, and, more especially, in its much more strongly punctured pronotum; it is also allied to catoxantha Lap.-Gory, dif- fering essentially in having the elytral vitta proceed from the humeral angles and not from the "milieu de leur base." The names dilatata and crenaia Mots., (Ancylocheira) , quoted by LeConte (Proc. Ac. Phil., 1873, P- 33^)^ appear to be unpublished; at any rate, I have carefully searched through four large volumes of Motschulsky's almost complete octavo papers without being able to find them, and they do not appear in the Munich Catalogue or in the special catalogue of the Buprestidas by Kerremans. Ornata, of Walker, is wholly unrec- ognizable but may be a subspecies of the very insufficiently described langi, of Mannerheim. There can be no doubt whatever of the reality of many species allied both to langi and gibbsi, although the pale antenna? of the latter constitute a very exceptional character. Langi is probably a local species and the common California form, with broadly sinuato-truncate and acutely bidentate elytral apices, named crenata in the table, is undoubtedly different. The species described by Harris nndcv ihervdiVCids Buprestis gcrafiii and charaderistica, belong respectively to the genera Acma-odcra and Chrysohothris. Buprestis aurulenta, of Linne, which iias gi\cn rise to so much trouble and uncertainty, having been identified with decora, lauta and hnpedita, of the following subgenera, has seemingly at length found a definite resting place as a European species and may therefore be removed from our lists. Subgenus Cypriacis nov. The body here becomes, as a rule, rather more oblong, with the pro- thorax somewhat less trapezoidal than in Buprestis proper, and the STUDIES IN THE AMERICAN BUPRESTIDiE II 7 pronotum uniformly impressed or sulcate along the middle. The labrum is more chitinous and metallic in its basal parts, but, aside from this, there is very little in the structure of the sclerites or organs of the body to distinguish it as a separate group. The complete absence of true elytral striie and the substitution therefor of a few longitudinal ribs, separated by wide and uniformly punctured or rugulose intervales, is evidently a result of advanced evolution through Buprestid species of the langi type, the latter having evenly and closely sulcate elytra, with the sulci densely and confusedly punctato-rugose; and the impressed median line of the pronotum, incipiently evident in those forms, becomes here still more developed as an external feature. We have only to imagine the coarse sulci of langi to be greatly expanded, coincidently with the suppression of the alternate ridges, to evolve the typical forms of this group, traceable indeed in intermediate stages through such species as adjecia, brevis and intricata. But the true striation of typical Bupresiis has so completely disappeared that the group would seem to require separation as a subgenus, and, if the intermediate forms above mentioned, which are however much more closely related to typical Cypriacis, as exemplified in lauta, than to any species of true Buprestis, were to become extinct, the differences would certainly be of full generic import. The species are generally of vivid green or bluish-green metallic coloration, with abruptly cupreous elytral side margins as a rule, though occasionally dull coppery-brown throughout. Those known thus far may be characterized as follows: — Elytral costae convex, their summits polished and impunctate. Pacific regions to Lake Superior 2 Elytral costae flat, punctured. Atlantic regions 13 2 — Intervales between the costae each with a feeble secondary ridge, that between the sutural ridge and the first normal costa strong and sub- similar to the latter 3 Intervales uniformly concave, densely and uniformly sculptured throughout . 5 3 — Elytral suture not at all cupreous. Body short, stout and convex, mod- erately shining, green, the head, lateral and apical parts of the prono- tum and side margins of the elytra cupreous; remainder of the elytra, less noticeable sublaterally and basally, with dark purplish-blue reflec- tion; under surface throughout of a bright metallic magenta; head strongly but rather sparsely punctate, more finely, densely so basally; eyes moderate, the antennae slender, dark piceous; prothorax strongly Il8 CASEY trapezoidal, with evenly and moderately arcuate sides, about twice as wide as long, the surface very moderately impressed along the median line, strongly, not densely punctate, coarsely and in part subconflu- ently so laterally; base gradually prominent medially; elytra three- fourths longer than wide, scarcely wider than the prothorax, the sutural and next two ridges strong, separated by narrow sulci, the intermediate ridges elsewhere low and smooth, the depressed lines rugose; apices narrowly sinuate and bidenticulate; under surface minutely, sparsely punctate, the prosternum flat medially, entering the metasternum for a considerable distance, the first ventral convex ; legs slender, the basal joint of the hind tarsi equal in length to the next two. Length 13.8 mm.; width 5.8 mm. Vancouver Island ((Nanaimo), — H. F. Wick- ham brevis n. sp. Elytral suture cupreous or subseneous, as well as the side margins 4 4 — Body rather stout, above bright brassy-green, densely punctured, the under surface bright coppery tinged with green, moderately punctate, the prosternum broadly impressed between the coxae and not hairy; head uniformly punctured; prothorax "almost three times as wide as long " ( !) , narrowed anteriorly, broadly rounded at the sides, cuprascent , broadly canaliculate; elytra at the sides broadly, at the suture narrowly, cupreous, each with four discal costae and one, scutellar, united mth the sutural, all smooth and shining, also with another costa, almost entire, between the elevated suture and the normal first discal costa; interstitial concave lines densely punctate, the apices subemarginate. Length 15.0 mm.; width 6.3 mm. Washington State (near Fort Van- couver) ad jecia Lee Body oblong-suboval, rather feebly convex, moderately shining, metallic green, the central parts of the pronotum and a broad median part of each elytron nearly from base to apex, deep indigo-blue, the elytral suture narrowly cuprascent, shading into green, the side margin more broadly brassy-green becoming cupreous posteriorly; under surface pale cupreous, with feebler greenish reflection; head very densely punctured throughout, with some small irregular smooth areas, the eyes not prominent; antennae black, with more or less cupreous reflec- tion; prothorax not quite twice as wide as long, the sides moderately converging and nearly straight to well before the middle, then strongly rounding and convergent to the apex; surface barely at all impressed along the median line, broadly, longitudinally impressed near each side, coarsely punctured, very densely almost throughout laterally; elytra about twice as long as wide, scarcely wider than the prothorax, the sides parallel, rounding and converging in rather less than apical third, the apices broadly rounded, not at all sinuate, the sutural angle right, not prominent; juxta-sutural supplemental ridge much narrower than the first normal costa though almost equally elevated; intermediate ridges low, narrowly smooth, the sublateral ridge very strongly elevated and conspicuous, the concave intervals strongly, transversely rugose; under surface finely, not densely but rather strongly i)unctate, the pro- sternum simply flattened medially, punctate anteriorly, more acuminate STUDIES IN THE AMERICAN BUPRESTID^ 119 posteriorly than in brcvis but scarcely at all entering the metasternum; first ventral strongly convex medially; legs short, slender, the first joint of the hind tarsi rather longer than the next two. Length 18.5 mm.; vi'idth 7.2 mm. New Mexico intricata n. sp. 5 — Third of the true elytral costae very short 6 Third costa long, extending to between apical third and sixth 7 6 — P^orm strongly depressed, the pronotum feebly impressed along the mid- dle. Body oblong, gradually attenuate posteriorly, above obscure and dark in color, the head and sides of the upper surface more or less cupreous, the bottoms of the elytral rugosities bluish-metallic; beneath not very shining, densely sculptured and cupreous; head finely, very densely punctured, with long and conspicuous pubescence, the eyes rather prominent; antennae nigro-metallescent; prothorax four-fifths wider than long, the sides rather strongly convergent and nearly straight, becoming arcuate and more strongly convergent only near the apex; base broadly and feebly arcuate, faintly sinuate laterally; surface con- vex, with moderate metallic punctures which are coalescent in small groups, dense laterally and also dense but smaller in the faint median depression; scutellum oblong, concave, cupreous; elytra nearly twice as long as wide, slightly wider than the prothorax, the parallel sides becoming gradually broadly rounded behind three-fifths, and thence feebly convergent and nearly straight to the rather narrow apices, which are rounded externally and not at all sinuate, the sutural angles right, not prolonged; surface notably depressed, with the usual convex costae strong, the third abbreviated a little behind the middle, the sub- lateral very much elevated and conspicuous, the intermediate surfaces coarsely, confusedly reticulato-rugose; under surface rather finely but densely punctate and conspicuously pubescent throughout, the pro- sternum broadly concave and densely sculptured and hairy, the first ventral convex medially. Length (c?) 13.4 mm.; width 5.3 mm. Lake Superior (Whitefish Point) lateralis n. sp. Form convex, the pronotum deeply canaliculate. Body rather broad, above obscure brassy-green; prothorax coarsely and densely punctate, narrowed anteriorly, the sides at the middle almost angulate, posteriorly sinuate; elytra densely and rugosely punctate, the suture, a scutellar costule, the margins and four discal costae on each strongly elevated and smooth, the third discal costa greatly abbreviated; apices some- what prolonged, truncate; under surface brassy-green. Length 15.5 mm. Lake Superior . sulcicollis Lee. 7— Under surface of the body duller cupreous, obscure or blackish toward the median line. Body large, oblong, moderately convex, the head and pronotum bright cupreous, the latter suffused with greenish on the central parts; elytra deep vivid green, the sutural ridges and the lateral margins, including the fourth costa, bright cupreous; head rather pubescent, coarsely, deeply punctate, the punctures longitudinally con- fluent and compressed, a small central region depressed and finely, densely punctured and more greenish; eyes not prominent; antennae I20 CASEY black; prothorax four- fifths wider than long, the sides feebly converg- ing and straight to about the middle, there gradually rounding and more convergent to the apex, the latter but feebly sinuate ; surface very coarsely punctured, sparsely so medially and sublaterally, more densely and confluently broadly near lateral fourth and near the sides, the median line distinctly impressed and finely, densely punctate; scutellum small, oblong; elytra not evidently wider than the prothorax, less than twice as long as wide, parallel, obliquely subrectiHnear at the sides in fully apical third, the apices broadly rounded, the sutural angles almost obtuse; sculpture as in lauta though somewhat coarser; under surface sparsely pubescent, finely, not closely punctate, the presternum broadly concave, strongly, sparsely punctate, not more pubescent, the concavity anterior becoming obsolescent posteriorly; apices of the ventral seg- ments cupreous-red throughout the width. Length 20. 8 mm.; width 8.0 mm. Mexico (Chihuahua), — C. H. T. Townsend. . *fabulosa n. sp. Under surface bright cupreous throughout 8 Under surface pale green, sometimes with faintly senescent reflection toward the median line, or bright bronze throughout 10 8 — Sides of the elytra posteriorly evenly rounding and converging to the apices. Body elongate-suboval, the elytra frequently feebly inflated subposteriorly, moderately convex, bright pale green to golden-green, sometimes darker greenish-blue, the anterior parts with more or less cupreous lustre, the elytral suture and side-margins bright-cupreous; head densely, confluently punctate, with a more depressed and more finely and densely punctured central spot which is longitudinally divided by a raised median line, subglabrous (9), or very much more finely and densely punctate throughout and conspicuously pubescent (d^); eyes larger and more prominent in the latter; prothorax nearly twice as wide as long, the sides nearly parallel as a rule to about the middle, there rounding and more converging to the apex, sometimes converging throughout; surface well impressed along the middle, moderately and not closely or confluently punctate medially, irregularly, densely so laterally; elytra three-fourths longer than wide, evidently wider than the prothorax, the apices broadly rounded, the sutural angle frequently minutely dentate; surface with the four discal ribs well developed, the intervales evenly and densely cribrate; under surface rather finely but strongly, not closely punctate, sparsely and feebly pubescent (9), or conspicuously so {d), the abdomen more closely punctured in the lat- ter; presternum more or less narrowly and distinctly impressed and punctured along the median line. Length 14. 0-18.0 mm.; width 5.6- 7.3 mm. Oregon to British Columbia and Vancouver Island. [=Ancy- locheira lauta Lee] lauta Lee. A — Similar to lauta in coloration and sculpture but smaller and narrower in form, the prothorax but little narrower than the elytra, the sides almost evenly converging and arcuate from base to apex; elytra and under surface as in lauta, except that the apices in the former are narrowly truncate; head in the male similarly densely ])unctured STUDIES IN THE AMERICAN BUPRESTIDiE 121 and hairy. Length 14. 0-15.2 mm.; width 5.6-6.0 mm. Oregon. [Fort Vancouver,— LeConte] radians Lcc. Sides of the elytra oblique posteriorly, becoming straight or feebly sinuate before the tips 9 g — Body rather stout, moderately convex, colored throughout as in laula; head (9 ) as in that species, e.xcept that the central more densely sculp- tured divided spot is not so clearly marked; prothorax four- fifths wider than long, the sides scarcely converging and nearly straight to well beyond the middle, there rounding and rapidly converging to the apex, sensibly incurvate at base; apex subtruncate, broadly bisinuate, the base also evenly bisinuate; surface with scarcely a vestige of impression along the middle but with a small impression before the scutellum, the punctures coarser than in laula and coalescent in short sinuous lines medially, coarser and densely coalescent laterally; scutellum broader, quadrate; elytra shorter, rapidly oblique at the sides in apical two-fifths, the oblique sides straight, the apices broadly rounded, becoming trans- verse suturally, the sutural angle right and rather blunt; surface as in lauta, the costae well elevated but with the intervales more unevenly though similarly densely cribrate; under surface finely, remotely punc- tate medially, rather strongly and closely so laterally, the prosternum flat and smooth, with an obsoletely subimpressed punctate median line. Length 15.3 mm.; width 6.0 mm. Colorado (Boulder Co.) . aBiiiulan. sp. Body much narrower, smaller than lauta but with similar coloration and sculpture, except that the green on the elytra is replaced by deep green- ish-blue; head (c?) similarly finely, very densely punctate and pubes- cent, the eyes larger and xnore prominent than in the female; prothorax as in lauta but with the sides becoming less parallel behind the middle though straight, not at all incurvate at base as they are in amula; punc- tures for the most part isolated or simply crowded as in lauta; elytra but little wider than the prothorax, as in lauta but about twice as long as wide, the sides becoming oblique in apical third and feebly sub- sinuate just before the apices, which are broadly subtruncate, the exter- nal angles right but rounded, the truncature anteriorly oblique sutur- ally, feebly undulated and with the sutural angle rather strongly spicu- late; sculpture as in lauta, except that the intervales are still more densely cribrate, almost -opaque; under surface as in lauta but more strongly and closely punctured than in the male of that species. Length (c?) 14-2 mm.; width 5.2 mm. Washington State (Tacoma) . tacomae n. sp. Body rather stouter, coloration nearly as in lauta throughout, excepting a broad bluish-green streak on each elytron; head nearly as in cemula; prothorax three-fourths wider than long, the sides converging and straight from the base nearly to apical third, there rounding and more convergent to the apex; surface narrowly impressed and finely, densely punctate on the median line anteriorly; punctures isolated, rather sparse, moderate in size, coarse and unevenly confluent laterally; elytra evidently wider than the prothorax, the converging sides posteriorly 122 CASEY becoming feebly sinuate before the tips, which are as in tacomcB, but with the outer angle more broadly rounded and more even and trans- verse suturally; costae much more feeble than in any of the preceding forms, becoming very low at a little before the centre of each elytron, where the punctures extend almost across them, the interv^ales densely, subevenly cribrate as in lauta; under surface punctured nearly as in lauta but more coarsely and densely so toward the sides, the prosternum rather narrowly and deeply impressed along the median line. Length ( 9 ) 16.5 mm.; width 6.2 mm. Idaho (Coeur d' Alene) . . nupta n. sp. 10 — Pronotum with coarse punctures coalescent in short isolated irregular series on the median parts, densely but irregularly confluent laterally; general coloration of the upper surface as in lauta 11 Pronotum with much smaller but deep perforate punctures, not evidently confluent in short series medially, very dense laterally; coloration of the upper surface dull cupreous, the elytra medially broadly cupreo-viri- descent 12 II — Body oblong, moderately convex, rather stout, the anterior parts more or less cupreous; elytra with bright cupreous inner and outer margins, the remainder bluish-green, the ribs frequently deep indigo; head with longitudinal rugulation caused by the compressed coalescent punc- tures, with a central green divided spot of dense punctures; eyes not at all prominent ( $ ), pale in all examples at hand; prothorax two-thirds wider than long, the sides broadly arcuate, becoming parallel basally, more convergent anteriorly, sometimes slightly sinuate just behind the apex, the latter truncate, broadly and evenly bisinuate; surface broadly and distinctly impressed along the middle; elytra evidently inflated behind the middle and nearly one-half wider than the prothorax to evidently less, scarcely three-fourths longer than wide, the oblique sides posteriorly becoming almost straight, the apices broadly rounded, transverse toward the suture, where the angle is sometimes distinctly spiculose; surface sculptured nearly as in lauta; under surface more strongly punctured, usually coarsely and very densely toward the sides ; prosternum deeply impressed and densely punctate along the median line; females abundant; males rare as usual in this group. Length (9) 16. 5-17. 5 mm.; width 6.5-7.0 mm. California venusta n. sp. Body larger and much more elongate, rather more convex, deep green, the anterior parts with but Httle cupreous lustre, the inner and outer mar- gins of the elytra cupreous; head as in the preceding but with the me- dian line more strongly elevated centrally; eyes not at all prominent, rather small; antennae black, metallic basally; prothorax two-thirds wider than long, the sides parallel and subsinuate in nearly basal half, thence moderately convergent and almost straight to the apex, which is truncate, with the sinuations obsolescent; surface broadly, moderately impressed and scarcely more finely though more thickly punctured along the middle, rather pubescent at the sides; elytra almost twice as long as wide, evidently wider than the prothorax, the converging sides poste- riorly not becoming quite straight, the apices very broadly rounded, STUDIES IN THE AMERICAN BUPRESTID.E 1 23 becoming somewhat oblique and straight toward the minutely denticu- late sutural angles; surface with the costa: rather more acutely elevated than usual, the concave intervales much more coarsely, confusedly and densely cribrate than in lauta; under surface plentifully pubescent, the presternum narrowly and deeply sulcate, the abdomen not closely but rather strongly punctate, more strongly but not densely so laterally, the general color pale green, aenescent along the median Une. Length 19.5 mm.; width 7.3 mm. New Mexico prospera n. sp. 12— Form elongate, subparallel, rather convex; under surface bright bronze, broadly greenish laterally except on the abdomen; head closely, perforately punctate, the punctures compressed and partially confluent longitudinally, the central densely punctate spot ill-defined, divided by an elevated line; eyes moderate, pale, only slightly prominent; prothorax only three-fifths wider than long, the sides converging and broadly, subevenly arcuate from base to apex; surface scarcely at all impressed along the middle but noticeably so at the base, the punctures much coarser laterally than medially, though not very confluent; scutellura subquadrate, rather well developed; elytra four-fifths longer than wide, a third wider than the prothorax, not at all inflated behind the middle but rather widest across the humeral swellings, the sides very gradu- ally rounding in apical two-fifths, becoming not quite straight at the apices, which are clearly truncate though with the outer angle obtuse, the sutural angles denticulate; costae rather low and broadly convex, the intervales with the dense and crowded punctures regular but unusu- ally deep; prosternum impressed and strongly punctate medially; abdomen finely, remotely punctate, becoming coarsely, shallowly and very densely so laterally. Length ( 9 ) 18.0 mm.; width 6.5 mm. Ore- tron afflnis n. sp. Form nearly similar, the coloration similar, except that the under surface is less bronzed and the median parts of the elytra more bluish-green; head more finely and densely punctate, the punctures compressed, the median line fine, elevated; eyes nearly similar; prothorax shorter, two- thirds wider than long, similar in form, the apex bisinuously truncate; surface distinctly impressed anteriorly and more finely, densely punc- tate along the middle but scarcely impressed posteriorly, the punctures generally close-set, extremely dense laterally, with a less punctate longitudinal area basally near lateral fifth or sixth; scutellum much smaller, oblong; elytra a little shorter, the sides more rapidly and arcuately converging behind, becoming subsinuate before the angles, which are rounded, the apices thence transverse to the denticulate sutural angles ; surface sculpture nearly as in the preceding but with more elevated costs, the cribration rather less coarse; under surface more strongly, less sparsely punctate, the punctures coarse and densely crowded laterally; prosternum finely, deeply grooved along the middle, else- where medially impunctate as usual. Length (9) 16.5 mm.; width 6.0 mm. California (Yreka) adulans n. sp. 12— Body rather stout and shorter, of less sombre coloration, the median parts of the elytra always more or less greenish or bluish. Obscure I 24 CASEY cupreous, the anterior parts with greenish patches, the elytra with the broad sutural and marginal cupreous stripes suffusedly and gradually disappearing internally, sometimes obliterating the medial blue or green area by the resulting greenish-aeneous tint; under surface cu- preous; head with densely compressed and longitudinally confluent punctures, rather coarse (?) or fine (cJ^), with a divided and more densely sculptured central spot, the surface feebly pubescent (9), or densely and conspicuously so ( d^) ; prothorax two-thirds to four-fifths wider than long, trapezoidal, the sides but slightly arcuate; surface only feebly impressed along the median line, strongly, sparsely punc- tured medially and in a sublateral tumescent area toward base, elsewhere densely punctate; elytra but little wider than the prothorax, sometimes decidedly inflated behind the middle, two-thirds or more longer than wide, the sides oblique posteriorly, the apices broadly rounded, becom- ing transverse toward the feebly denticulate sutural angles; surface with the low flat costae distinctly and rather closely but irregularly punctate, the intervales very densely and evenly cribrate; under sur- face pubescent and rather closely punctured (;j^), or subglabrous and more sparsely and finely punctate (9), the prosternum impressed and punctured medially. Length 14. 3-17.0 mm. ; width 5.3-7.0 mm. New York, Ontario, Michigan and Lake Superior (Bayfield and Whitefish Point) impedita Say Body much more elongate and more parallel as a rule, always dark coppery- brown, the median line of the elytra seldom wdth any decided greenish tinge 14 14 — Elytra gradually narrowed from but slightly behind the middle to the apices, which are transversely truncate. Body elongate-suboval, moderately convex, rather shining, dark brown with feeble cupreous-red lustre, more pronounced on the elytral costae, the under surface bright cupreous; head ( 9 ) as in the preceding, the eyes only moderate in size and but slightly convex; prothorax two-thirds wider than long, strongly trapezoidal, the sides broadly arcuate, becoming subsinuate broadly toward base, the basal angles somewhat acute from above; surface only very obsoletely impressed though more densely punctate along the middle; punctures coarse and deep, disposed as in the preceding; elytra fully four-fifths longer than wide, behind the middle two-fifths wider than the prothorax, the external angles at apex obtuse but not rounded, the sutural scarcely denticulate; surface with low flat punc- tured costa}, the intervales more coarsely and confusedly cribrate than in the preceding, the second and third decidedly greenish in tinge; under surface evidently though sparsely pubescent, finely, sparsely punctate, more strongly but not densely so toward the sides of the abdomen, the prosternum impressed and closely punctured along the middle. Length (9) 17.2 mm.; width 6.4 mm. Ontario (Georgian Bay). canadonsis n. sp. Elytra more abruptly narrowed behind in barely apical third, the sides scarcely becoming straight near the apices; intervales of the elytra never greenish 15 STUDIES IN THE AMERICAN BUPRESTID.C 1 25 15 — Elytral apices nninded. Body oblong, rather more convex, sometimes feebly inllated beliind the middle, slightly shining, dark coppery-brown, duller laterally than medially; head as usual in the preceding species, with dense compressed punctures and fine central carina; prothorax three-fourths or more wider than long, the sides straighter and more parallel in basal, more converging in apical, half, sculptured as in the preceding but more closely throughout, the apex similarly feebly bisin- uously truncate; scutellum elongate, subquadrate, impressed, cu])reous; elytra three-fourths or more longer than wide, from very evidently to at least two-fifths wider than the prothorax, the apex transverse at the suture, broadly rounded externally, the sutural angles right; surface with the low flat costae punctured, closely and coarsely so externally, the intervales very densely, not very coarsely cribrate; under surface bright cupreous, the sculpture nearly as in canadensis. Length 1 3.0-18.0 mm.; width 4.8-6.8 mm. Pennsylvania to Ontario .. obscura n. sp. Elytral apices narrower, truncate to sinuato-truncate and more or less evidently bidenticulate. Body more elongate and parallel, relatively narrower and somewhat less convex, dark coppery-brown to blackish; head a little larger than in obscura, with dense compressed punctures which are rather coarse and subglabrous ( 9), or finer and moderately pubescent (c?), with a median densely punctate depressed area divided by the central carina and generally green in color; eyes only moderate in size but prominent, generally black; prothorax almost twice as wide as long, the sides as a rule strongly converging in apical, nearly parallel and straighter in basal, half, the surface scarcely at all or very feebly impressed along the middle, the sculpture as in canadensis but closer throughout, the lustre generally more cupreous laterally; scutellum broader, cupreous, concave; elytra four- fifths longer than wide, but slightly wider than the prothorax, seldom at all inflated behind the middle, the sculpture nearly as in obscura though rather coarser; under surface not cupreous but bronzed, the abdomen rather closely punctate throughout and somewhat more strongly so than in obscura. Length 13. 5-18. 8 mm.; width 4.7-6.7 mm. North Carolina, Alabama and Louisiana striata Fabr. Several of the above species are represented by series sufficiently full to show clearly the extent of specific variation. This is perhaps most evident in the form of the prothorax, which may be subpromi- nently rounded at the sides to evenly trapezoidal with feebly and evenly arcuate sides, within specific limits, also in the degree of in- flation of the elytra, which are often notably wider behind the middle than at the humeri. The form of the elytral apices, on the other hand, seems to be one of the more constant characters, and, on the whole, the most useful in characterizing certain species which resemble each other superficially. The species described by Say under the 126 CASEY name impedita was wrongfully united with striata by LeConte; it is an abundantly valid species. Subgenus Stereosa nov. This section of Buprestis is very limited in extent. The general habitus of the species, involving among other features the somewhat greater convexity of the body, is quite different from that prevailing in either of the preceding subgenera, and it is this that prompts me to give them a distinctive subgeneric name. The general anatomical characters are as in Cypriacis, but the elytra have no vestige of regular discal costce and but feeble traces of impressed striae, in the usual sense, but instead there are series of large perforate punctures, the general surface being closely, deeply and confusedly but evenly crib- rate with smaller punctures. The prothorax is trapezoidal and sel- dom has any trace of the impressed median line so developed in Cypriacis and in some of the less typical forms of true Buprestis, but the ante-scutellar impression or deep puncture is generally evident. The basal joint of the hind tarsi is similarly variable in elongation. The four species known thus far may be described as follows: — Elytra narrowly but abruptly sinuato- truncate and feebly bidenticulate at apex. Body narrowly elongate-oval, convex, moderately shining, greenish-subcupreous, the elytra generally bright green, with the suture and external margins cupreous, frequently with a broad and ill-defined median vitta on each of a dense indigo blue and again at times suf- fused with cupreous almost throughout; under surface vivid green to bright cupreous; head very densely compresso-punctate, the central dense spot not very constant; eyes rather small and not very prominent; antennae of the usual form, black; prothorax two-thirds or more wider than long, the sides converging from base to apex and nearly straight; apex truncate, very obsoletely bisinuate, the base broadly but more evidently; surface strongly but not closely punctate, the punctures smaller and denser in patches sublaterally, a median impunctate line sometimes evident but never entire; elytra slightly wider than the pro- thorax, almost twice as long as wide, the sides rounding and converging in about apical third; surface subevenly convex, the small punctures not densely crowded, rather sparse suturally, the larger perforations of the series moderate but variable in size and well separated; under surface finely, closely punctured, sparsely along the middle, the punc- tures toward the sides of the abdomen not at all crowded ; prosternum flattened, finely, densely punctate and sometimes feebly impressed; head and under surface but slightly more hairy in the male. Length studip:s in the American buprestid^ 127 (c?) ir.3-14.6, (9) 15. 2-16. 7 mm.; width (c?) 3.9-5.0, (9) 5.2-6.0 mm. North Carolina to Arkansas and Louisiana. [ == salisburyensis Weber] decora Fabr. Elytra rounded at tip, the sutural angles acute and more or less prominent. 2 2 — Coloration nearly as in the preceding, bright and metallic, with the elytral suture and side margins cupreous. Form shorter and relatively stouter than in decora, generally paler and more uniform brassy-green, the elytra sometimes broadly suffused with darker, though usually greenish-blue; under surface bright cupreous; head smaller; similarly sculptured though more coarsely in the female, where there are some irregular smooth spots; prothorax similar but more rounded at the sides and more longitudinally convex, sometimes vdth vestiges of an impressed and more punctate median line, the punctures still coarser, not dense and more evenly disposed; elytra only two-thirds or more longer than wide, slightly wider than the prothorax, almost similarly formed, the serial perforations coarser, the punctures similar though relatively smaller and more apt to be distributed in a single line on some of the intervals, owing to the coarser serial punctures; under surface as in decora, the relative hairiness of these parts and of the head of the male and female similar. Length 1 2.0-1 3.6 mm.; width 4.5-5.5 mm. New Jersey and New York ultramarina Say Coloration nearly uniform and more sombre, the elytra without cupreous suture or margins 3 3 — Form oblong-oval, moderately convex, dull or feebly shining, obscure cupreous-brown, the anterior parts brighter but with the more elevated portions black (9), or with the elytra obscure green (c?); under sur- face bright cupreous throughout, with greenish reflections; head very densely compresso-punctate, more coarsely so and less hairy in the female, the median line carinate centrally; eyes moderate, not very prominent; prothorax trapezoidal with feebly arcuate sides, three- fifths wider than long, with a rather pronounced groove along the apical margin laterally; punctures strong, sparse medially, where there is a nearly entire impunctate line in the female but not in the single male at hand, dense and confluent near the sides; scutellum oblong, depressed on the disk, with a median elevation; elytra three-fourths or more longer than wide, evidently vdder than the prothorax, the oblique sides in apical third almost straight (c?), to decidedly arcuate through- out (9), the surface even, except a sublateral ridge which is obsolete basally and feeble (J^), or very pronounced ( 9 ), the serial perforations rather coarse, more or less elongate, the cribration dense, becoming very crowded laterally; under surface finely punctate, sparsely so (9), or more densely and more pubescent (cJ*). Length 16.0-18.0 mm.; width 5.8-7.0 mm. Alabama (Grand Bay) cribripennis n. sp_ Form more elongate and rather more convex, dark and uniform bronze- brown above, more cuprascent anteriorly, bright cupreous throughout beneath; head more densely and uniformly compresso-punctate, the eyes decidedly smaller and less convex, not at all prominent; prothorax Proc. Wash. Acad. Sci., April, 1909. 128 CASEY similar in form but rather more closely punctate, the punctures similarly- uneven in distribution; scutellum nearly similar, cupreous; elytra nearly twice as long as wide, rather coarsely but not quite so densely cribrate, more notably sparsely so toward the suture, the punctures of the series coarse, deep and perforate; sublateral ridge less prominent; under sur face somewhat more strongly and less sparsely punctate, more strongly but not very closely so laterally; male and female apparently with but slight structural differences. Length 18.0-22.0 mm.; width 7.0-7.8 mm. North Carolina; also one specimen labeled " Cal. " but no doubt erroneously. [ = bosci Lap.- Gory] apricans Hbst. The sublateral ridge of the elytra varies in prominence in this sec- tion very much as in Cypriacis, where it is very strong and imparts a very characteristic facies in such species as lateralis and intricata and feeble in fabulosa and brevis; here it is almost obsolete in decora but pronounced in apricans and especially in the female of cribripennis. This ridge is seldom very prominent in Buprestis proper, though generally traceable. All the species of this section are locally distrib- uted in pine regions and therefore generally rare, though abundant enough within their proper environments, Ultramarina seldom seems to have a coloration wholly warranting its name, the bright surface generally being of a more golden-green color than in decora; it seems to be decidedly rare. Dicerca Esch. This genus, though so different in general habitus, is related closely to Buprestis in most of its general organization, excepting in antennal structure and the very much shorter basal joint of the hind tarsi; it differs otherwise principally in its rough and irregular surface sculp- ture, prolongation of the elytral apices, which however becomes obso- lete or nearly so in some species and occurs sporadically in other parts of the family, as for instance in Psiloptera and in Buprestis consularis, and, finally, in the male and female characters at the apex of the abdomen, which, though of a markedly different kind, are similarly inconstant within s])ccific limits and therefore unreliable as a specific criterion. The broadly and feebly sinuato-truncate and sub- lobate anterior margin of the prosternum is a rather more striking feature than in Buprestis, but the prosternal process, the mesosternum and the metasternum arc virtually similar and the mcso-metasternal suture is very fine, the two surfaces being similarly anchyloscd and not STUDIES IN THE AMERICAN BUPRESTID^ 1 29 departing noticeably in structure from the corresponding parts in Psiloptera and Gyascutus, as we are led to suppose from the group divisions made by LeConte in his monograph of the American species of the family. The male does not differ much from the female in superficial char- acters, but the sterna are apt to be more impressed and punctate and more hairy in the former, where the apex of the fifth ventral is trans- versely, rectangularly emarginate as a rule, with the median part of the notch briefly and broadly produced in a truncate lobe. In the female the apex is more acuminate and is narrowly and deeply emargi- nate, generally with a slender tooth projecting from the bottom of the sinus, sometimes as far posteriorly as the lateral lobes; this tooth may, however, be well developed to wholly obsolete within specific limits and the sinus itself may be broader or narrower. In the male the intermediate tibiee generally have on the inner side an obtuse tooth, which occasionally becomes a slender erect process as in horni. The coloration of the body is brassy or bronzed to dull cupreous as a rule, but is sometimes deep black throughout and without metallic lustre of any kind. The species are very numerous in subarctic North America and are assignable to a considerable number of distinct groups, for the most part recognized and clearly characterized by LeConte; these groups may be briefly indicated as follows: — Prosternum flattened, concave or sulcata 2 Presternum convex 10 2 — Prothorax not constricted basally or arcuately inflated before the middle 3 Prothorax shorter and more transverse as a rule, sinuously narrowed basally behind the more or less marked anterior inflation, the pronotum gen- erally more uneven in sculpture 7 3 — Intermediate tibiae not dentate in the male though sometimes obtusely swollen or subangulate within 4 Intermediate tibiae of the male with a strong obtuse internal tooth 5 Intermediate tibiae of the male with a slender erect serrulate process inter- nally 6 4 — Elytra obliquely and gradually narrowed posteriorly, the sides usually becoming feebly sinuate before the apices, which, however, are not prolonged; they are deeply sinuate and acutely bidenticulate; pronotum never obviously canaliculate Group I 130 CASEY Elytra with the tips similarly sinuate and bidenticulate but prolonged; pronotum deeply canaliculate Group II 5 — Elytra as in Group II but with the greatly prolonged apices entire. Group III 6 — Elytra more strongly, arcuately narrowed behind, the apices rather abruptly but only very briefly prolonged and broadly obtuse. Group IV 7 — Prosternum and last ventral segment not bicostate; elytral apices mod- erately prolonged 8 Prosternum with two smooth and strongly elevated costse, the fifth ventral bicostulate 9 8 — Elytral apices entire Group V Elytral apices sinuate and acutely bidenticulate; front with a callous band between the eyes Group VI 9 — Elytral apices markedly prolonged, rounded; metasternum strongly sulcate Group VII 10 — Body in form and sculpture somewhat as in the obscura group; elytra obliquely attenuate, not prolonged, the tips obtuse; intermediate tibiae not dentate Group VIII Although these groups differ among themselves to a very noticeable extent in general appearance and sculpture^ I am not prepared at present to assign any very great taxonomic weight to their differential characters. The convex prosternum of mutica Lee, constituting group VIII of the table, seems at first very important, but in compar- ing Buprestis confluenta and Iceviventris, corresponding differences are observable in even more marked degree. It will be noted that I have been compelled to make some changes in the grouping proposed by LeConte. Group I — Type ohscura Fabr. The species of this group are numerous and require unusual care in discrimination. The hurried and superficial manner in which they have been hitherto surveyed may be suflSciently inferred from the statement made by Crotch (Pr. Acad. Phil., 1873, p. 85) regarding his inability to distinguish lurida from ohscura, two species altogether different, not only in general appearance, form of the prothorax and relative prominence of the eyes, but by the very much more marked internal swelling of the middle tibiae in the male of the former; they are in fact representative species of the two principal sections into which this group may be divided, as shown by the following table: — Eyes prominent; intermediate tibiae of the male not or only very feebly swollen internally 2 STUDIES IN THE AMERICAN BUPRESTID^ I3I Eyes not at all prominent, the intermediate male tibiae more or less distinctly swollen or angulate internally; prothorax sinuate at apex throughout. . .9 2 — Prothorax evenly narrowed from base to apex, the latter sinuate as usual 2 Prothorax parallel, the sides rounding and converging anteriorly to the apex 4 3 — Surface feebly convex, very roughly sculptured; body broad, oblong- oval, rather shining, dull cupreous, with the elevations black; under surface blackish and obscure cupreous; head roughly sculptured, with uneven callosities, which tend to form a transverse band between the eyes, the latter only moderately prominent (9), or strongly so (d^); prothorax one-half to three-fourths wider than long, the sides feebly arcuate, sometimes feebly subprominent before the middle; surface finely, sparsely punctate, with a more strongly impressed medial sub- apical area, coarsely and confluently punctate laterally, with sublateral line of broken callous spots and an impression near lateral fifth behind the middle; ante-scutellar cavity distinct, single; scutellum small; elytra scarcely wider than the base of the prothorax, three-fourths or more longer than wide, the oblique sides posteriorly becoming evidently sinuate before the tips, which are more broadly sinuate than in any other species of the group, strongly bispiculose; surface densely punctate, sparsely and finely near the suture, generally vdth evident rows of coarser punctures, the alternate intervals with elongated and con- spicuous irregular raised polished black areas; under surface coarsely, rugosely punctate, the pectoral channel much broader, more evenly concave and densely sculptured and hairy in the male than in the female as usual. Length 13.5-18.3 mm.; width 4.7-6.6 mm. Virginia (Nor- folk) and North Carolina (Southern Pines) to Florida. [ = Buprestis obscura Fabr., baliimorensis Hbst., and consimilis and pruinosa Gory]. ob.«cura Fabr. Surface evidently more convex and smoother, form more elongate, still darker in coloration, with more aeneous than cupreous lustre; head coarsely, very densely but more evenly punctato-rugose, without very marked callous spots, the pubescence longer and much more conspicu- ous; eyes more widely separated; prothorax nearly similar in form but rather less transverse, the sides evenly and feebly arcuate throughout; surface much more even and more evenly punctate, rather closely so, with similar anterior punctured area and basal pit medially, the punc- tures relatively less coarse and more evenly dense and confluent later- ally, not impressed sublaterally behind the middle; scutellum larger; elytra similar though rather more elongate, the sculpture of the same general nature but less coarse and more even, the callous spots smaller, less elevated and much less conspicuous; sides posteriorly becoming straight and not sinuate toward the tips, which are much more narrowly and normally sinuate and bidenticulate; under surface nearly similar. Length (?) 18.8 mm.; width 6.3 mm. Texas. indurata n. sp. 132 CASEY 4 — Apex of the prothorax subtruncate, broadly sinuate laterally, the median part arcuate, with a small median sinuation. Body very much smaller, oblong, feebly inflated behind the middle, rather convex, moderately shining, dull aeneous above, more cuprascent beneath; head densely, moderately coarsely punctato-rugose, with a transverse callous spot centrally, the pubescence short, not conspicuous; antennas cupreous, sparsely hairy; prothorax three-fifths wider than long, the parallel sides nearly straight to wdthin an unusually short distance of the apex, there rather abruptly rounded and convergent to the apex ; surface con- fluently and coarsely punctate laterally, less coarsely, rather sparsely medially, where the median line is feebly impressed before the middle and rather more punctured, the ante-scutellar pit bipunctate; near lateral fourth before the middle and at base there is a small irregular callous area and behind the middle a large vague subobUque impressed region; scutellum very small; elytra with the basal margin slightly wider than the prothorax, not quite twice as long as wide, the grad- ually rounding obhque sides in apical two-fifths becoming barely per- ceptibly sinuate near the emarginate tips, which are strongly bispicu- lose; surface densely and unevenly punctate, less densely so suturally, the impressed striae distinct suturally but only finely punctate, the raised elongate spots, separating the usual subdepressed areas of denser and finer punctuation, moderately developed; under surface rugosely punctate, the median tooth of the apical sinus ( 9 ) unusually broad and parabolic in form, the two notches very small. Length ( $ ) 14.0 mm.; vddth 5.0 mm. Wisconsin truncata n. sp. Apex of the prothorax broadly, subevenly sinuate as usual 5 5 — Sides of the elytra posteriorly very gradually rounding and oblique, becoming straight or barely perceptibly and very broadly sinuate before the emarginate apices ; body small in size 6 Sides of the elytra more rapidly converging posteriorly, becoming very obviously sinuate before the apices; body larger and stouter. 8 6 — Elytra unusually elongate, more than twice as long as wide, with the punctures everywhere distinct, close but not confluent laterally. Body slender, convex, shining, feebly cupreous above and beneath, the ele- vations blackish; head coarsely, densely punctato-rugose, with a trans- verse central callus, the pubescence evident; prothorax one-half wider than long, the sides straight and parallel to apical third, there rounding and convergent to the apex; surface somewhat tumid medially, feebly, obliquely impressed near the base at about lateral fourth, with coarse crowded punctures and some small and very irregular callous spots laterally, the median parts sparsely but strongly punctured, the ante- scutellar pit not evident; elytra at base very slightly wider than the pro- thorax; surface unevenly and feebly striate, not so roughly sculptured as usual, the elongated callous spots only slightly elevated, the areas of dense punctures well defined and conspicuous; under surface coarsely punctato-rugose, the [)ectus ( 9 ) narrowly and deeply, the first ven- tral unusually distinctly but broadly, sulcate, the fifth ventral with a STUDIES IN THE AMERICAN BUPRESTID.E I33 short blunt median tooth at apex. Length ( 9) 15.7 mm.; width 4.9 mm. New Hampshire (Hampton) gracilis n. sp. Elytra less elongate, twice as long as wide or less, with the coarse punctures laterally very uneven and confluent as usual 7 7 — Body oblong, moderately stout and convex, sometimes feebly inflated behind the middle, moderately shining, obscure aeneous, somewhat cupreous beneath; head nearly as in gracilis, the prothorax similar but shorter and more transverse, with the parallel sides in basal two- thirds generally, straight but sometimes faintly sinuous; surface not evidently tumescent medially and with the ante-scutellar pit evident, sometimes bipunctate; elytra narrowing behind in about apical third, roughly sculptured, with the patches of dense punctures large and very obvious, the elongate callous spots strong; under surface ( $ ) nearly similar, with the impression of the first ventral rather distinct, the median tooth of the apical sinus longer and acute or finely aciculate; male with the sternal channels broader, concave, densely sculptured and more hairy. Length 13. 8-16.0 ram.; width 4.8-5.7 mm. Kansas to Pennsylvania porcatula n. sp. Body narrower anteriorly, the elytra rather strongly inflated, convex, nearly similar in coloration, lustre and sculpture, except that the densely punctate patches of the elytra are less impressed, more confused and less definite; head nearly similar though much smaller; prothorax three-fifths wider than long, slightly inflated before the middle, with broadly rounded sides, the latter evidently sinuate in nearly basal half, the basal angles less than right; elytra behind the middle fully a fourth wider than the prothorax, twice as long as wide, the sides broadly arcuate, gradually converging behind in fully apical two-fifths, the apices narrower than in porcatula and more semicircularly sinuate; female with distinct sternal and abdominal axial impressions, the fifth ventral with a small shallow apical sinus, the median tooth and lateral lobes very short, the former obtuse. Length (?) 14.0 mm.; width 4.7 mm. Florida (Marion Co.) innocua n. sp. 8 — Form oblong-oval, convex, shining, obscure aeneous, brighter cupreo- aeneous beneath ; head coarsely, densely punctato-rugose, vsdth a small transverse central callous spot, the hairs very small and inconspicuous; eyes unusually large and prominent; antennae but little longer than the head, cupreous; prothorax large, but little more than one-half wider than long, the sides broadly rounding and converging anteriorly, becoming parallel from before the middle and also evidently sinuate basally; surface rather closely, moderately punctate along the middle, especially in the anterior feebly impressed area, the punctures gradually coarser and close laterally, becoming coarse and crowded near the sides, vdth a sublateral line of irregular callus but not evidently impressed near the base, the ante-scutellar pit small though deep; elytra but little vnder than the prothorax, fully twice as long as wide, faintly inflated subposteriorly, narrowing in apical third; surface not very rough, with coarse and very confluent irregular punctures laterally, without evident series of punctures, the punctured patches moderate, the callous spots 134 CASEY large though feeble, cariniform near the semicircularly emarginate apices ; under surface ( 9 ) deeply impressed throughout the sterna and first ventral, the sinus of the fiJfth deep, with the median tooth long, very slender and aciculate. Length (9) i6.8 mm.; width 5.9 mm. Indiana sagax n. sp. Form nearly similar, shining, obscure to bright cupreous above, not brighter beneath ; head nearly similar but without so distinct a central callus, sub- glabrous, the eyes relatively not quite so large or prominent though more so than usual; prothorax similar but broadly rounding at the sides and more gradually narrowing anteriorly from only a little before the middle, the sides parallel and subsinuate behind the middle; surface sculptured as in sagax but with the medial feebly impressed and densely punctate subapical spot shorter and better defined, and with large diffuse obhque sublateral impressions of the disk behind the middle; scutellum sim- ilarly distinct, rounded; elytra very evidently more than t\\ice as long as wide, not so inflated behind the middle, more strongly though sim- ilarly sculptured, the densely punctate areas deeper, the callous spots stronger, especially in the sublateral series behind the middle; there are also more or less evident, scarcely impressed series of coarser punctures; under surface ( 9 ) nearly similar, except that the medial tooth of the apical sinus, though variable in length, is truncate at tip, sometimes broadly so. Length (9) 17.5 mm.; width 6.0-6.3 ^^- Florida. florid£B n. sp. 9 — Oblique sides of the elytra posteriorly becoming straight or virtually so toward the apices. Grayish-brassy in color; prothorax shorter than wide, narrowed anteriorly, the sides broadly rounded, subparallel behind the middle, the surface roughly punctate, confluently toward the sides, sparsely at the middle, with an obsolete and nearly smooth vitta near each side, the median line somewhat impressed, more densely punctate, the channel being more evident than in ohscura; elytra rugosely punctate, the striae toward the suture and apex more strongly punctured, the intervals variegated with irregular elevated black spaces, more irregular than in ohscura, being rather portions left between confluent punctures than oblong areas, the apices gradually attenuate, not at all prolonged, bidentate. Length 15.0 mm. (9). Pennsylvania. soror Lee. A — Cupreous, the elevations greenish-black; under surface obscure cupreous-red; head coarsely, very densely punctate, with callous spots in a general broad transverse line between the eyes and above the middle of the front; prothorax three-fifths wider than long, the sides straight and parallel, gradually rounding and converging to the apex from a little before the middle, the surface almost evenly but only moderately convex, coarsely, densely punctured laterally, with numerous smooth, very irregular raised spaces, more finely, sparsely so medially, with the densely punctate median line impressed ante- riorly but not elsewhere; ante-scutellar pit diffused on the margin; elytra more than twice as long as wide, but little wider than the pro- thorax, gradually narrowing behind in fully apical two-fifths, the STUDIES IN THE AMERICAN BUPRESTID.E I35 convex surface more even than usual, with impressed punctured stria? almost even and visible nearly to the sides, the densely punc- tured spots small and not impressed, the black spots feebly elevated, generally small but occasionally large and oblong, the punctuation not very rugose though dense, except near the suture; under surface ( 9 ) channeled along the axial line as usual, the apical sinus with a long slender aciculate tooth. Length (9) 16.0 mm.; width 5.8 mm. New York (Buffalo) regularis n. subsp. Oblique sides of the elytra toward tip becoming more or less evidently sinuate, the tips thus appearing as though very slightly prolonged. ... 10 10 — Elytra not noticeably inflated subposteriorly, relatively more elongate. 1 1 Elytra more or less conspicuously inflated, being decidedly wider behind the middle than at base 12 II — Form elongate, rather convex, moderately shining, obscure aeneous, feebly cuprascent beneath; head coarsely, densely punctate and rugu- lose, the central callus obvious; prothorax nearly two-thirds wider than long, the sides rounding and converging anteriorly, parallel and broadly, feebly sinuate from before the middle to the base; surface subevenly convex, coarsely, densely punctured laterally, with callous spots near the sides medially and others, very variable, in a sublateral line especially visible before the middle and at base; median parts rather finely, sparsely punctate, the anterior punctured area but very feebly impressed, the basal pit evident, sometimes bipunctate; scutel- lum longitudinally sulcate as a rule; elytra but little wider than the prothorax, distinctly more than twice as long as wide, subparallel, the sides gradually rounding and converging in apical two-fifths and gradually sinuate near the strongly bidenticulate apices, which seem to be slightly prolonged; surface coarsely, very rugosely and confluently punctate laterally, the punctured series broken but evident, the raised spaces very irregular and the punctate areolae not well defined; under surface as usual, the median tooth in the apical sinus of the fifth ven- tral (9) long and acutely aciculate. Length (9) 17. 5-18.2 mm.; width 5.7-6.1 mm. Indiana and Illinois (northern) .... levettei n. sp Form narrower and still more convex, glittering and bright aeneous, varied with small subcupreous patches, the raised parts greenish-black; under surface cupreous, dull laterally due to the coarse confluent sculpture, polished and viridiaeneous medially; head very coarsely, confluently punctato-rugose, with small callous spots; antennae a little longer than the head, bright cupreo-aeneous; prothorax shorter, three-fourths wider than long, the sides broadly arcuate and gradually moderately converg- ing anteriorly, becoming subparallel and feebly sinuate only behind the middle; surface very coarsely, densely punctate and with numerous very irregular ramifying callous spots laterally, sparsely and more finely punctate medially and nearly as in levettei; elytra more than twice as long as wide, barely at all wider than the prothorax, the sides very gradually rounding and converging from only slightly behind the middle, more feebly and broadly sinuate toward the strongly bispicu- 136 CASEY lose apices than in the preceding; surface less rugosely though coarsely and rather confluently punctate laterally, the punctured stria; fine but evident suturally, the dark spots elongate and numerous but only feebly elevated, the punctured areolae very irregular but distinct, some- times with more greenish-metallic lustre; under surface (9) deeply sulcate along the prosternum, the metasternum flat, the medial tooth of the apical sinus long, slender and aciculate. Length ( 9 ) 16.0 mm.; width 5.5 mm. Illinois (southern) gaudens n. sp. 12 — Body elongate and rather narrow, convex, gradually inflated posteriorly, slightly shining, dark cupreous, obscure above, brighter when living; head rather small, densely, moderately coarsely punctured, generally with a V-shaped central callus; prothorax two-thirds wader than long, the sides broadly rounded and moderately converging before, becom- ing parallel and broadly sinuate behind, the middle; surface broadly, indefinitely impressed sublaterally behind the middle, densely but only moderately coarsely punctate and with irregular callous spots laterally, sparsely and finely punctate medially, the subapical spot scarcely impressed, the basal pit variable, generally small but sometimes double; elytra at base but little wider than the prothorax, gradually broader posteriorly, and, behind the middle, evidently wider than at base, the sides thence very gradually rounding and converging, becoming sinuate before the sharply bidenticulate apices; surface unusually rough, the cupreous areolae more deeply impressed, the raised spots numerous and prominent; under surface deeply impressed along the axial line nearly to the apex of the first ventral. Length 15. 0-17.0 mm.; width 4.9-5.8 mm. New York to Florida. [ =indistincta Mels.] . . lurida Fabr. Body nearly as in lurida in general coloration and moderate lustre but more parallel and relatively stouter, less dilated posteriorly; head less densely punctate, the callous spots similarly very irregular in a transverse inter- ocular region; prothorax larger and much less transverse, barely one- half wider than long, similar otherwise, except that the anterior densely punctate medial spot is more impressed and the areas of dense punctua- tion at each side of the sublateral line of callus anteriorly are more impressed, causing the vitttc to be more prominent; elytra more rapidly attenuate posteriorly, scarcely more than twice as long as wide, but slightly inflated behind the middle, the surface very much smoother than in lurida and with the punctured series very distinct and entire nearly to lateral third; sides only slightly sinuate before the apices; under surface ( 9 ) nearly similar, except that the sinus of the fifth ventral is shallower, with the median tooth shorter, triangular. Length (9) 17.0 mm ; width 5.9 mm. Texas ser'ata n. sp. Body shorter than in either of the preceding, shining, xneous, the head and sides of the pronotum cuprascent, the entire under surface more bril- liant cupreous; head coarsely, densely punctate, with a broadly V- shaped central callus, including a depressed area; there is also a trans- verse band of irregular callous spots between the antcnnx; prothorax nearly as in lurida but with the oblicjue impressions beliind tlie middle still stronger; elytra scarcely twice as long as wide, moderately inflated STUDIES IN THE AMERICAN BUPRESTID^ 1 37 behind the middle, where they are a third wider than the prothorax, the sides rather rapidly rounding and converging posteriorly, becoming only broadly and very moderately sinuate near the apices; surface very densely but not very coarsely punctato-rugose laterally, very un- even, the punctured areolae much impressed, the striae indistinct and much broken, the callous areas moderately elevated, finely cariniform near the apices; under surface ( 9) as usual, the median tooth of the apical sinus long and slender but obtuse at apex. Length ( 9) 15.0 mm.; width 5.5 mm. Iowa ero.sa n. sp. Obscura is represented before me by a large series, holding very homogeneously to the characters above stated; cabinet specimens are generally dark and obscure in coloration, but the appearance of fresh examples is markedly different, owing to a pale grayish-metallic bloom which is destroyed unless the utmost care be taken in handling them; it is probably one of these specimens that served as the type of pruinosa. The integuments of the body are so thick and hard that a pin is to be inserted only by using great force; lurida and allied species are not so heavily armored. Soror is compared with obscura by LeConte, and is declared to be identical by Crotch, though the form of the prothorax seems to approximate more closely to that of the lurida type; nothing is stated by the describer regarding the relative size or prominence of the eyes, and I have therefore had to assume that the species will enter the second section of the group. It is entirely possible, however, that this is a mistake and that it is more related to porcalula, in which event regularis will become a true species. Group II — Type pugionata Germ. This group agrees with the last in nearly all of its characters, includ- ing the unmodified middle tibiae of the male and the deeply sinuate and acutely bidenticulate elytral apices, but here, the latter are more prolonged and the pronotum differs in having a pronounced median impressed channel and subentire smooth sublateral callous vittas, imparting a distinctly different habitus to this part of the body. In pugionata, the elytral tips are as much prolonged as in the next group. The only two known species are the following: — Form elongate, the surface golden-brassy in lustre; head densely punctate, with a V-shaped central callus and two on the occiput separated by a moderate channel; prothorax shorter than wide, moderately narrowed anteriorly, the sides anteriorly broadly rounded, the surface conflu- 138 CASEY ently punctate, trisulcate and marked with four smooth elevated vittae; elytra densely punctate, the sutural stria distinct, the suture elevated; alternate intervals with elongate smooth elevated spaces, the sides posteriorly oblique and gradually converging [becoming sinuate], the tips slightly prolonged and bidentate; under surface coppery, densely and coarsely punctured at the sides, nearly smooth or sparsely punc- tured at the middle; prosternum deeply, the first ventral moderately, sulcate, the metasternum almost flat; male with the last ventral tri- dentate, the medial tooth short and broadly truncate, the middle tibiae straight and unmodified; female with the medial tooth of the apical sinus acute. Length 15.0 mm. Alabama to Pennsylvania. [==pugio- naia Lap.-Gory nee Germ.] lepida Lee. Form moderately elongate, much smaller, convex, shining, somewhat ob- scure coppery above, brighter cupreous beneath; head rather small, very coarsely but not so densely punctato-rugose, with large irregular central callous spaces; eyes rather small but prominent; prothorax short, three-fourths wider than long, the sides feebly arcuate and mod- erately converging before the middle, subparallel and feebly sinuate basally, the apex feebly sinuate; surface with a densely punctured median channel, with the ridges broad, nearly flat and sparsely punc- tate, thence to the sides coarsely, confluently punctate, with a more or less incomplete and narrower callous vitta, and some smaller callous spaces still nearer the sides; elytra barely wider than the prothorax, much more than twice as long as wide, the sides gradually rounding and con- verging from shghtly behind the middle, becoming broadly sinuate, the tips greatly produced as in the next group, deeply sinuate and bispicu- lose, the surface uneven, with series of small punctures suturally, rather coarsely, densely and unevenly punctate laterally, the alternate intervals with strongly elevated but very irregular, elongate and smooth greenish-black spaces, the finely punctured areola? rather distinct; under surface axially grooved, more strongly in the male, in which the middle tibia? are unmodified, the median lobe of the apical emargi- nation moderately broad, truncate, the medial tooth of the female sinus long and slender. Length 11. 0-12.0 mm.; width 3.6-4.0 mm. Massachusetts. [=Buprestis pugionata Germ, (nee Lap.-Gory)]. pugionata Germ. In regard to lepida Lee, which I have been unable to examine, it is stated by Crotch (1. c. p. 87) that the prothorax is longer than wide, but, though perhaps unusually elongate, it certainly cannot even be so long as wide, as indeed definitely so stated by LeConte. Group III — Type divaricata Say. In this group the elytra are more prolonged at apex than in any other, this character giving the species a habitus which is altogether distinctive. It is the largest group of the genus and comprises a wide STUDIES IN THE AMERICAN BUPRESTID^ 1 39 range of sculpture and outline, so that there ought not to be much difficulty in recognizing the various forms, which are in great part as yet undescribed, I have restored two of Melsheimer's species, be- lieving them to be valid, almost as distinctly so in fact as prolongala, of LeConte, but caudata, of that author, which has been viewed with suspicion, is one of the more aberrant forms of the group, indicating the quality of observational work hitherto put upon them. There may be some mistakes in my interpretations, owing to paucity of material, but the large series of divaricataheiore me, gives a very good idea of such variation as may legitimately be anticipated, so that I do not think there will prove to be any considerable amount of actual synonymy. The species and subspecies may be defined as follows: — Body of an elongate-oval, convex form, the elytra gradually rounding and converging behind from only slightly behind the middle; coloration generally cupreous above and beneath; species confined apparently to the extreme northeastern regions and not at present known to occur to the westward of New England 2 Body oblong-elongate and more parallel, the elytra rounding less gradually at the sides posteriorly and from a point always well behind the middle. 8 2 — Elytral prolongation more or less abruptly formed, with its sides becom- ing parallel for some distance before the apex 3 Elytral prolongation gradually formed and more acuminate, not parallel even at the tip ; species very small in size 7 3 — Prothorax broadly angulate or angularly rounded at the sides well behind the middle 4 Prothorax obtusely subprominent at about the middle of the sides, the latter not sharply angulate 6 4 — Series of punctures on the elytra not evident even toward the suture. Body moderately elongate, shining, dark and obscure in color above, the punctures and depressions deep cupreous; under surface and legs obscure cupreous; head coarsely, densely punctato-rugose, the central callus small and inconspicuous; eyes relatively small, only moderately convex and not noticeably prominent; prothorax three-fifths wider than long, the sides very feebly convergent and sinuate for two-fifths, there broadly but rather sharply angulate and thence more convergent and nearly straight to the apex, which is broadly sinuate as usual; surface sparsely punctate medially, impressed and more punctured on the me- dian line to behind the middle, coarsely and densely so, with numerous uneven callous spots irregularly disposed, broadly toward the sides, with two strong deep punctures before the scutellum; elytra slightly wider than the prothorax, more than twice as long as wide, strongly, unevenly punctured, loosely suturally, very densely laterally, with numerous small and moderately elevated, very irregular and ramify- I40 CASEY ing black areas, the slightly depressed cupreous areolae small, well de- fined, finely and closely punctate; prolongation cupreous; under sur- face and femora coarsely, closely and rugosely punctate, the axial line smoother and impressed as usual, the fifth ventral ( $ ) binigrocostulate medially, the median tooth of the apical sinus acute but broad at base, triangular. Length (9) 16.0 mm.; width 5.4 mm. Massachusetts. abrupta n. sp. Series of fine punctures, which are rather remotely spaced, rather distinct on the elytra except laterally, the series unimpressed 5 5 — Body short and rather stout, convex, only moderately shining, cupreous, with obscure or blackish elevations, the under surface dark cupreous; head coarsely, densely punctate, canaliculate posteriorly, with numer- ous longitudinal rugae centrally, the eyes moderate in size but very con- vex and extremely prominent, deep black in the type; prothorax fully three-fourths wider than long, the sides parallel and deeply sinuate for a little more than a third, there broadly but very sharply angulate and thence strongly converging and straight to the apex; surface broadly impressed and more closely punctate throughout the length on the median line, with a large basal impression containing two deep punctures, coarsely, deeply, subconfluently punctate and with scattered irregular callous spots laterally, feebly impressed sublaterally behind the middle; elytra nearly as in the preceding but shorter, only a little more than twice as long as wide, the surface uneven, irregularly sculptured, very densely laterally, the raised spots feebler, the cupreous areola? almost similar, the prolongation shorter and broader; under surface nearly similar, the black ridges of the fifth ventral feebler and less approxi- mate. Length (c^) 15.0 mm.; width 5.2 mm. New Hampshire (Hampton) biangulata n. sp. Body elongate-oval, convex, shining, dull cupreous, the elytra obscure greenish throughout though varied by feeble subamescent reflections, the prolongation not different in color; under surface dark cupreous; head coarsely, unevenly rugose and punctate, with a depression before the central callus; eyes rather well developed, somewhat prominent; prothorax barely three-fifths wider than long, the sides subangularly prominent slightly behind the middle, thence subparallel and feebly sinuate to the base and converging and broadly arcuate to the apex; surface canaliculate and densely punctured before the middle and simply more closely punctate posteriorly on the median fine, the basal pit large, deep and single, the sculpture otherwise nearly as in the preceding; elytra aberrant, similar to abrupta in form though more gradually attenuate, but almost uniformly punctate, without densely punctate cupreous areolae and with the raised spots feeble, small and few in number; punctured series unimpressed and scarcely traceable, except near the suture; under surface nearly as in abrupta. Length (9) 17.0 mm.; width 5.8 mm. Massachusetts iiiflatula n. sp. 6 — Form elongate-oval, convex, rather shining, cupreous with the eleva- tions black, brighter cupreous beneath; head coarsely, very densely but subevenly punctato- rugose, the eyes moderately convex and rather STUDIES IN THE AMERICAN BUPRESTID^ I4I prominent; prothorax fully three-fourths wider than long, the sides obtusely but subangularly prominent at the middle, thence sul parallel and feebly sinuate to the base and converging and feebly arcuate to the apex, the median line obsoletely impressed, deeply so anteriorly and with a very feeble elongate central callus, densely punctate ante- riorly, the basal fovea deep, single or double; punctures laterally coarse, deep and confluent, with few irregular callous spots near the sides but with more forming an irregular broken vitta at lateral fourth, the im- pressions obsolete; elytra scarcely wider than the prothorax (c?), or slightly wider ( 9 ), much more than twice as long as wide, the prolong- ation generally very abruptly formed and parallel, broader than usual though rather longer than wide; surface strongly, almost uniformly punctate, densely and rugosely at the sides, with numerous callous spots, which are elongate and regular suturally but elsewhere small, very irregular and ramifying transversely and longitudinally, the areolae not well defined and not more finely punctate; under surface as usual. Length 14. 2-19.0 mm.; width 4.7-6.4 mm. Massachusetts. pisciformis n. sp. A — Similar to the preceding but narrower, paler and brighter cupreous, the elytra with the elevations more uniformly elongate and regular throughout and the sides more gradually converging posteriorly, with the prolongations much less abrupt, more extended and very notice- ably narrower in form, brighter cupreous in color; eyes more prom- inent, especially in the male. Length 16. 5-1 7.2 mm.; width 5.3-5.8 mm. Massachusetts longicauda n. subsp. 7 — Body moderately stout, convex, shining, cupreous with black elevations, dark cupreous beneath; head moderately coarsely, very densely and subevenly punctate, with a small transverse central callus, the eyes rather small, moderately prominent; prothorax two-thirds wider than long, the sides converging and broadly rounded anteriorly, gradually becoming parallel and sinuate behind the middle; surface feebly im- pressed and more punctured along the middle, deeply and more nar- rowly impressed anteriorly, the two ante-scutellar punctures conspicuous ; punctures laterally only moderately coarse, crowded but not confluent, wth callous spaces and sublateral broken vitta nearly as in the preced- ing, the impression obsolete; elytra at base wider than the prothorax, the humeral angles well exposed; sides parallel, gradually rounding and converging posteriorly from about the middle, becoming broadly, feebly sinuate to the tip of the unusually acuminate prolongations, the latter strongly carinate medially on the disk; surface much as in the preceding species, the series of fine punctures readily traceable suturally; under surface as usual. Length (d^) 13.2 mm.; width 4.5 mm. Massachu- setts cupreola n. sp, A — Similar to the preceding but narrower and more elongate, the head a little smaller, more coarsely, confusedly punctate, with the ramify- ing interspaces broader and more ruguliform; eyes nearly similar but scarcely so prominent; prothorax shorter and broader, not im- pressed along the median line except anteriorly and more coarsely and 142 CASEY confusedly, densely punctured laterally, with the callous spots more diffused; elytra at base much less evidently wider than the prothorax, the sides still more gradually converging posteriorly, and, from a little behind the middle, the surface nearly similar but with the black eleva- tions more numerous, irregular and confused, except near the suture, the small and clearly defined, cupreous areolae deeper. Length ( d^) 13.0 mm.; width 4.3 mm. Massachusetts filiola n. subsp 8 —Prothorax narrowed gradually from base to apex. Form moderately stout, convex, shining, cupreous with the smooth parts blackish above, rather dull cupreous beneath; head confusedly and densely punctato- rugose, more or less impressed along the median line throughout, with a central interrupting callus; eyes moderately prominent; prothorax at base three-fourths wider than long, with the sides slightly prominent behind the middle; surface feebly impressed and closely punctate along the median line, more deeply anteriorly, the two basal punctures strong; punctures laterally only moderately coarse, very dense on each side of the sublateral ragged line of callus, coarser and very dense with much mingled callus near the sides; elytra not evidently wider than the base of the prothorax, twice as long as wide to the base of the cauda, the sides gradually rounding and converging in apical two-fifths, the prolongations gradual, becoming parallel and strongly divaricate; surface densely, rugosely punctured laterally, finely, sparsely so sutur- ally, without distinct punctured series except feebly near the suture and without distinct finely punctured areolae, the black spots only feebly elevated, elongate suturally but small and irregularly ramifying else- where; female with the medial terminal tooth short, broad and triangu- lar. Length (9) 18.4 mm.; width 6.2 mm. Pennsylvania. caudata Lee. Prothorax subparallel and generally slightly sinuate at the sides behind the middle 9 Prothorax evidently though gradually narrowed basally, the width at the base more or less obviously less than at the middle, where the sides are more or less inflated though rounded 19 9 — Sides of the prothorax obtusely but sharply angulate at the middle. Body stout, convex, partially shining, black with the punctures green anteriorly, the elytra, except suturally and throughout the width bas- ally, dull cupreous and green intermingled, the elevations black as usual; under surface in great part black, the punctures cupreous; head confusedly but not very densely punctate, with numerous small iso- lated callous sjwts which are very irregular in form, the eyes rather small but prominent; prothorax three-fourths wider than long, the sides smuate in basal, converging and nearly straight in a[)ical, half, sj)arsely punctate medially, with a small apical punctate impression and a closely double basal pit, the punctures laterally not very coarse and everywhere isolated, mingled with much black callous area; surface unimpressed; elytra not wider than the prothorax, between two and three times as long as wide, the sides gradually rounding and con- verging in posterior two-fifths, the prolongation gradually formed but STUDIES IN THE AMERICAN BUPRESTID^ I43 long and becoming parallel apically, ol)tuse, each with a strong, an- teriorly bifurcating carina; surface almost even, with but very few scarcely elevated and almost evenly oblong smooth spots, strongly, coalescently punctate laterally and with even, unimpressed series of punctures, nine in number, only the extreme lateral series being indis- tinct; under surface rugose as usual, deeply sulcate along the entire axial line to the tip of the first ventral. Length (c?) 20.0 mm.; width 6.4 mm. Pennsylvania (Allegheny) subaequalis n. sp. Sides of the prothorax sometimes slightly prominent medially, but always rounded from a dorsal viewpoint 10 10 — Body above deep and uniform blue-black throughout, moderately slen- der and convex, rather dull in lustre; head very densely, uniformly punctato-rugulose, impressed along the median line throughout; eyes prominent, moderate in size; antennas piceous-black basally, not metal- lic; prothorax three-fourths wider than long, the sides angularly prom- inent at the middle, parallel and feebly sinuate to the base, moderately converging and barely arcuate to the apex; surface with a large oval impressed and punctate area before the middle, flattened and less punc- tate thence to the deep single basal pit, laterally coarsely and very densely punctate, obliquely, deeply impressed behind the middle, with a small prominent callus before the middle and another at base in the same sublateral line, also a small submedian callus near the lateral margin; elytra slightly wider than the prothorax, between two and three times as long as wide, the sides gradually rounding and converging in apical two-fifths, rather abruptly sinuate at the base of the prolonga- tions, which are short, broad and parallel, separated by a large acute notch ; surface almost even, coarsely punctured except near the suture, very densely laterally, the striae very distinct and impressed to beyond the median, line, the callous spots small and very few in number, feeble, the areolae few in number, and, behind the middle, large, shallow, less coarsely but not at all finely punctate and not metallic; under surface very deeply sulcate along the axial line. Length (J*) 18.5 mm.; width 6.0 mm. New York (Adirondack Mts.), — J. F. Kemp. . nigra n. sp. Body above metallic aeneous, bright to obscure, sometimes moderately cuprascent 11 1 1 — Sides of the elytra more strongly rounding and converging behind, the prolongations rather abruptly formed and short, parallel 12 Sides of the elytra more gradually rounding and converging through a longer distance posteriorly, the prolongations very gradually formed, longer and more acuminate 15 12 — Elytral striae strong and distinct, sometimes nearly to the sides but not sulciform 13 Elytral striae very fine, sometimes traceable only to the median line 14 13 — Body rather stout, moderately convex, shining, aeneo-cupreous, the under surface slightly more intense in color; head coarsely, very densely punctato-rugose, with sparse callous spots, more or less impressed along the median line; eyes only moderately prominent; antennae Proc. Wash. Acad. Sci., April, 1909. 144 CASEY Eeneo-cupreous ; prothorax four- fifths wider than long, the sides obtusely subprominent at the middle, thence strongly converging and evidently arcuate anteriorly, and parallel and broadly sinuate basally; surface rather more punctured but not or only feebly impressed along the mid- dle, except the constant and distinct impressed spot near the apex, the basal pit single or double, toward the sides strongly, obliquely im- pressed behind the middle, the impression generally deep and con- spicuous, the punctures coarse and dense, with the usual callous spots very irregular; elytra nearly as in nigra in outline, the prolongation not quite as long as wide in its parallel part as a rule, obtuse at the tips but with the sutural angle sometimes acutely produced; surface rather rough, coarsely, densely punctured laterally, finely and sparsely near the suture, the striae rather coarse, not much impressed but visible sometimes throughout the width, strongly punctured; callous spots strong, elongate suturally, smaller and ramifying elsewhere ; areolae not very well defined and not much less coarsely punctured ; under surface as usual, the axial sulcus stronger in the male, and, on the prosternum, broader, more concave and more pubescent; medial apical tooth .(9) short, broad and triangular but variable as usual. Length 17.0-20.5 mm.; width 5.6-6.8 mm. New York, Pennsylvania, Indiana, Illinois, Kentucky and Wisconsin (Bayfield). [=Buprestis divaricata Say and D. dubia Mels. ; also Bnprestis acuminata Lap.-Gory nee Pall.( acu- minata Pall., is European)] divaricata Say A — Similar in general form and size but very dark and obscure aeneous, obscurely cuprascent beneath, more shining, the head more coarsely rugose, the eyes a little larger and more prominent, the pronotum more coarsely punctured laterally, with larger and very irregular smooth spaces and the impressions obsolete, the median parts except anteriorly very remotely punctate, not at all impressed behind the middle, the basal pit similar; elytra with the punctures everj'where well separated and almost to the side margins, where, however, they become more or less transversely coalescent, the striae obvious and impressed. Length (9) 19.0 mm.; width 6.1 mm. New York (Buffalo) IJmula n. subsp. B — Similar in general form but smaller and more slender, rather less bright and darker aeneous, the under surface cupreous, \nth much metaUic green lustre medially; head with the depressed medial spot bright green; eyes similar; prothorax similar but with the sides ob- tusely angulate at the middle, the converging sides thence to the apex straight; surface less impressed behind the middle, with the sculpture toward the sides much coarser; elytra similar but more elongate, the surface very much more coarsely and contluentlypunc- tato-rugose, except near the suture, the impressed stria: evident almost throughout the width, the black smooth spots small but strongly elevated, widely separated; two ridges of the fifth ventral strong as usual in this group. Length (9) 16.5 mm.; width 5.2 mm. In- diana incisa n. subsp. Body less stout and more convex, shining, dark aeneous with black spots as STUDIES IN THE AMERICAN BUPRESTID^ I45 usual, more cupreous and greenish beneath; head nearly as in divari- cata but with the eyes relatively a little larger and less widely separated; prothorax similar but with the sides anteriorly more convergent and nearly straight, posteriorly much less sinuate, the sculpture more even and with the post-median impressions much more feeble; elytra nearly as in divaricata throughout but vdth the prolongation, though similarly abrupt, still shorter, the parallel part nearly twice as wide as long; punctuation distinct and clear-cut but as usual coarse and confluent laterally, the stria; less coarse though very distinct to beyond the median line; under surface and sexual characters similar, except that the two ridges of the fifth ventral are shorter, more apical and very closely approximate, separated only by a narrow groove, — a very variable fea- ture however. Length (9) 17.0 mm.; width 5.8 mm. New York (Buffalo) parumpunctata Mels. 14 — Form rather stout and convex, rough in sculpture, moderately shining, largely aeneo-cupreous above with many small green clouds in the de- pressed areas of the elytra, the entire frontal margins and labrum also bright green; under surface cupreous and bright green, the legs green with feeble cupreous reflection; head very coarsely rugose, more finely punctato-rugose anteriorly, the eyes only moderately convex and not prominent; prothorax nearly as in divaricata but less closely sculptured throughout, obliquely though moderately biimpressed behind the mid- dle; elytra but little wider than the prothorax, almost as in divaricata in outline and in the broad, obtuse and rather abruptly formed pro- longation, but differing in its fine and finely punctate striation which is not visible much beyond the median line, the black callous spots numerous and strongly elevated, the depressed areolae large, confused, with the close punctures coarse but smaller than the others; under surface rugose, nearly smooth and impressed along the axial line as usual but not obviously less rugose along the middle of the abdomen; median abdominal tooth ( 9 ) rather long and slender. Length ( 9 ) 18.2 mm.; width 6.2 mm. Ohio aurichaicea Mels. Form somewhat similar but smaller and with relatively smaller head and prothorax, darker and more uniform aeneo-cupreous above and but moderately shining, cupreous with some green glints beneath ; headless coarsely, densely punctato-rugose, with a better defined and depressed, more finely punctate central spot, which is viridi-aeneous as well as the labrum and the median apical parts of the front; eyes moderately large and more prominent; prothorax three-fourths wider than long, the sides subparallel, very broadly arcuate, gradually moderately converg- ing anteriorly and becoming sinuate in only basal third, the surface less coarsely rugose, though closely punctured laterally and similarly unimpressed along the middle except very feebly anteriorly, the basal puncture small and deep, the oblique impressions diffuse and moder- ate; elytra evidently vdder than the prothorax, similar in outline but with the obtuse prolongation narrower and relatively a little longer, though with the parallel part not longer than wide, similarly subabruptly formed; surface more uniform and less rugose, the fine and finely punc- 146 CASEY tured striae visible to some extent throughout the width, the lateral margins much more brilliantly metallic aeneous and green than the remainder of the disk, the elevations moderate, the areolae not well defined and not densely punctate, except those behind the middle; under surface similarly rugose and sulcate,the prosternal sulcus broadly concave but only moderately pubescent in the male. Length ( (^) 15.8 mm.; width 5.2 mm. Vermont (Bennington Co.) . . . aestiva n. sp. 15 — Elytral striae fine. Body rather narrow, convex and shining, obscure aeneo-cupreous, the under surface brighter, cupreous; head densely and confusedly compresso-rugose, with a broad entire median impressed line, the eyes moderately prominent; prothorax fully four-fifths wider than long, the sides rather prominently rounded at the middle, thence converging and feebly arcuate to the apex and parallel and broadly sinuate to the base; surface not impressed medially, except narrowly and rather deeply anteriorly, the basal pit deep, only very widely and diffusely subimpressed laterally behind the middle, coarsely, closely punctate and confusedly rugose; elytra sHghtly more than twice as long as wide, just visibly wider than the prothorax, the sides gradually rounding and converging in posterior two-fifths, becoming but very feebly and broadly sinuate, the prolongation rather short and broad but altogether gradual in formation, obtusely truncate, not becoming parallel at tip; surface coarsely and almost regularly punctured, coal- escently so toward the sides, the callous spots moderately convex, few in number and widely separated; areolae wanting; fifth ventral (9) closely bicostate, the median apical tooth small but acute. Length (?) 15.5 mm.; width 5.1 mm. New Hampshire (Hampton) . . rustican. sp. Elytral striae coarse 16 16 — Pronotum more or less evidently and generally deeply, obliquely impressed sublaterally behind the middle 17 Pronotum without oblique impressions behind the middle 18 17 — Form subparallel, moderately broad and convex, rather shining, ob- scure cupreous, darker beneath; head coarsely, confluently and very densely but somewhat evenly punctato-rugose, feebly depressed along the middle, the eyes only moderately prominent; prothorax nearly as in the preceding but with the sides more deeply sinuate basally, the base somewhat narrower than the middle; surface more or less impressed and closely punctate throughout the length medially, rather deeply, confusedly and coarsely punctato-rugose laterally; elytra slightly wider than the prothorax and evidently more than twice as long as wide, sometimes feebly inflated subposteriorly, the sides very gradually rounding and obliquely converging in apical two-fifths, the prolongation moderately wide, obtuse, gradually formed, the incisure distinct and deej:) but narrow; surface coarsely sulcate except laterally, the sulci coarsely punctate, the intervals convex, the sublateral interval posteriorly more ])romincnt than usual, the sculpture densely confused laterally, the callous spots moderate and rather few in number, the areoke not well defined except suturally and posteriorly; under surface as usual, the apical tooth ( 9 ) small but very aciculatc, sometimes wholly want- STUDIES IN THE AMERICAN BUPRESTIDiE I47 ing, the sinus then becoming narrower. Length 15.7-20.5 mm.; width 5.2-6.7 mm. Michigan, Indiana, New York (Adirondacks) and New Hampshire subcuprea n. sp. A — Similar to the preceding but in the male narrower and more parallel, the elytra not wider than the prothorax, narrowing more rapidly in about apical third and drawn out at apex to a narrower but sim- ilarly gradual prolongation, the prosternal channel wider than in subcuprea (&), leaving much narrower lateral ridges, the meta- sternum also with an evidently broader and less posteriorly attenuated furrow. Length (c?) 17.7 mm.; width 5.4 mm. New Hampshire. pertinax n. subsp. Form slightly stouter, convex, shining, grayish-metallic, not at all cupreous above though bronzed beneath ; head nearly as in subcuprea but more impressed along the median line almost throughout, the eyes only moderately developed, somewhat prominent; prothorax less transverse though rather more than three-fourths wider than long, nearly similar in outline though much less sinuate at the sides basally; surface deeply but not very broadly sulcate along the median line, the sulcus more closely punctured as usual, very coarsely, extremely densely and con- fusedly punctured laterally, with an occasional callous spot, the oblique impressions not so deep as in the preceding; elytra somewhat more inflated, and, behind the middle, fully a fourth wider than the pro- thorax, the sides very gradually rounding behind, becoming similarly gradually and broadly sinuate to the tip of the moderate prolongation, which is more broadly obtuse at apex ; surface with rather coarse sub- sulciform series of coarse punctures, the sculpture dense and very con- fused laterally, the subsutural carina posteriorly sharply defined but elsewhere with very confused and irregular elevated spots and having but few well defined areolae; under surface as usual in this group, the medial apical tooth ( 9 ) very short and broad, triangular. Length (9) 20.5 mm.; width 6.8 mm. Colorado sulcatula n. sp. 18 — Body larger, stout and convex, moderately shining, obscure bronze, brighter beneath ; head very coarsely and confusedly rugose, with abroadly impressed median line interrupted centrally by a V-shaped callus; eyes well developed but rather feebly convex, scarcely at all prominent; prothorax nearly as in the preceding but relatively smaller, the sides behind the middle parallel and deeply sinuate nearly as in subcuprea, the surface not impressed though more plentifully punctate along the middle, except the usual subapical spot, the basal pit with sloping pos- terior surface, the sculpture laterally very much confused, being a mixture of coarse and dense punctures and large ramifying callous areas; elytra at base distinctly wider than the prothorax, the sides nearly straight, gradually rounding and converging in posterior two- fifths, the gradually formed cuprascent prolongation unusually short and broad, with the sutural angles minutely aciculate; surface with distinct close-set punctures which are only partially confused near the sides, the black callous spots numerous and strongly elevated, giving a coarsely rugose general surface, the series not much impressed but 148 CASEY very evident except laterally, coarsely punctured suturally, the areolae not well defined; under surface coarsely rugose, smoother and chan- neled axially as usual, the apex of the fifth ventral ( 9 ) with the sinus small, the lateral lobes very broad and the median tooth broadly tri- angular. Length ($) 20.0 mm.; width 6.7 mm. Vancouver Island (Victoria) vancouveri n. sp. Body rather small in size and less stout, moderately convex, shining, ob- scure brassy-bronze in color, brighter beneath; head densely, moder- ately coarsely and subevenly punctato-rugose, with a central depression before a small transverse callus; eyes rather small and not very prom- inent; prothorax fully four- fifths wider than long, the sides broadly, subevenly arcuate and converging from slightly behind the middle to the apex, parallel and sinuate in but little more than basal third; sur- face not or scarcely impressed medially, except narrowly so anteriorly, the basal pit triangular and deep, densely, coarsely and confusedly punctate laterally, with a few callous spots; elytra but slightly wider than the prothorax, parallel, gradually and broadly rounding and con- verging in apical two-fifths, the prolongation gradually formed, atten- uate and rather long; surface with rather coarse striae which are visible almost throughout, the striae only feebly impressed but coarsely and unusually distinctly punctured, thfe sculpture coarse, dense and con- fused laterally, the callous spots almost wanting before, moderately conspicuous behind, the middle, the areolae not defined; under surface rugose and axially sulcate, the last ventral ( J*) with the deep emargi- nation transversely rectilinear at the bottom, the lobe occupying the entire width. Length (c^) 15.5 mm.; width 5.0 mm. Wisconsin. angusticauda n. sp. 19 — Pronotum not impressed along the median line except anteriorly 20 Pronotum more or less deeply sulcate throughout along the middle 21 20 — Form rather narrow, elongate and parallel, moderately convex, shining, obscure gray- metallic in lustre, scarcely bronzed beneath; head coarsely, confusedly compresso-punctate, feebly but unequally impressed along the middle, the eyes well developed and prominent; prothorax nearly twice as wide as long, the sides prominent, almost obtusely angulate at the middle, where the width is but slightly greater than at base, thence gradually converging and feebly arcuate to the apex, broadly sinuate basally; surface obliquely impressed sublaterally behind the middle, coarsely, confusedly punctato-rugose laterally, with a definite callous spot only before and behind the impression; elytra nearly two and one- half times as long as vnde, barely wider than the prothorax, the sides very gradually rounding and converging in apical two-fifths, becoming gradually and broadly sinuate, the prolongation long, somewhat atten- uate and cuprascent, the surface unevenly sulcate, the elevations only moderate in number and prominence, the foveote rather small and definite posteriorly but not more finely punctate; under surface rugose and deeply sulcate axially, the last ventral ( 9 ) with the apical sinus but little wider than deep, the tooth gradually and very finely aciculate. Length (9) 19.0 mm.; width 6.0 mm. Wisconsin (Bayfield), — H. F. Wickham subargentea n. sp. STUDIES IN THE AMERICAN BUPRESTID.E I49 Form much stouter, convex, subinflated posteriorly, rather dull, blackish with the depressions very obscure reneous above, the under surface very ob- scure ffineo-cupreous and black; head densely compresso-rugose, im- pressed along the middle, the eyes smaller and only moderately prom- inent; prothorax strongly, subangularly prominent at the sides just before the middle, thence strongly, arcuately narrowing to the apex, less strongly so with feebly sinuate sides to the base, very nearly twice as wide as long, the surface deeply, obliquely impressed behind the middle sublaterally and also with a deep rounded impression before the middle at each side of the median tumid surface, coarsely, densely and confusedly punctato-rugose laterally; elytra slightly wider than the prothorax, the sides gradually rounding and then rather rapidly con- verging in apical two-fifths, the prolongation narrow, moderate in length but not abruptly formed; surface rugose laterally, coarsely punctato-sulcate thence suturally, the surface sublaterally with the elevations bordering the large areolae posteriorly unusually strong, giving a feebly undulated appearance; female with the terminal sinus narrow and as deep as wide, the tooth very small and broadly angulate, not extending posteriad through a fourth of its depth. Length ( 9 ) 18.5 mm. ; width 6.2 mm. Ontario severa n. sp. 21 — Thoracic sulcus narrow and better defined 22 Thoracic sulcus very broad, much dilated toward the base, less sharply defined at the sides and generally with a small central callus 24 22 — Elytra with only a series of levigated spaces near the margin. General characters nearly as in tenehrosa (!) except that the pronotum has no levigated elevations and the impressions, except the median channel, which is better defined, more obsolete; elytra distinctly furrowed, especially near the suture, the furrows punctured ; the attenuated apex of the elytra is longer than in tenebrosa, rather truncated and beneath is of a dark blue; the prosternum is linear and the base of the abdomen scarcely channeled; the teeth of the fifth ventral ( $ ) are longer than in tenebrosa and of a brilliant ruddy-copper. Length (?) 15. 5-18.0 mm, British America, (found in latitude 54° and at Cumberland House). [ =Stenuris tenebrka ELirby] tenebrica Kirby Elytra with levigated spaces almost throughout as usual 23 23 — Color gray-metallic, often pruinose [as usual when fresh]; prothorax more than t\vice as wide as long, tlie sides anteriorly strongly rounded, posteriorly subsinuate, punctate, canaliculate, at each side behind the middle obliquely and deeply impressed; elytra deeply striate, the inter- vals with more obscure oblong spaces which are smooth and rather elevated, posteriorly obliquely attenuate, prolonged, the apices rounded, scarcelyi divaricate ; beneath sparsely pubescent, axiallysulcateas usual; last ventral of the male tolerably deeply emarginate at apex, sinuate ;( $ ) with the lateral teeth rounded, the medial acute. Length 19.2-21.2 mm. Lake Superior; [also British America] prolongata Lee. A — Form more elongate and parallel, the elytra narrowing more rapidly behind in scarcely more than apical third, the prolongation long, 150 CASEY gradually formed and decidedly narrow; prothorax twice as wide as long, prominently inflated and rounded at about the middle, the sulcus evident but shallow, deeper anteriorly, the sublateral impres- sions deep, the surface laterally very confusedly but subevenly puncta- to- rugose ; elytra parallel, not wider than the prothorax, more elongate, coarsely sulcate, the sulci coarsely punctate, obliterated only near the sides, the elevations small and not conspicuous, the sublateral interval posteriorly slightly more prominent and more or less broken as usual; apical sinus ( 9 ) rather small, twice as vnde as deep, the tooth rather long, slender and aciculate. Length (c?, 9) 19. 0-21. 2 mm.; width 6.2-6.8 mm. British America longipennis n. subsp. Color obscure bronze, a little more coppery beneath though dark; body smaller, moderately convex, rather dull; head coarsely punctato-rugose, impressed along the middle, the eyes moderate, rather prominent; prothorax fully four-fifths wider than long, the sides evenly and broadly rounded, strongly converging anteriorly, feebly so and slightly sinuate basally; surface deeply sulcate, with a large deep basal pit, coarsely, densely and confusedly but almost evenly punctato-rugose laterally, the oblique impressions feeble but evident; elytra slightly wider than the prothorax, very gradually rounding and converging at the sides in apical two-fifths, the prolongation moderately wide, rather short and gradually formed, the median notch large and deep; surface very con- fusedly rugose throughout, the elevations numerous, small and irregu- lar in form, the sulci only indicated by vague Unciform impressions, which are coarsely punctate but only definitely so near the suture; pro- sternum of the female with an unusually broadly concave rugose im- pression, rapidly narrowing posteriorly, the metasternum broadly con- cave, the first ventral impressed only anteriorly; medial tooth of the apical sinus rapidly, very finely aciculate. Length (9) 16.0 mm.; width 5.6 mm. New York callosa n. sp. 24 — Form stout, rather convex, not very shining, dusky cupreous, brighter beneath; head coarsely, densely compresso-rugose, feebly impressed along the middle, the eyes moderate, prominent; prothorax nearly four-fifths wider than long, the sides prominently rounded rather be- hind the middle, thence strongly converging and slightly arcuate to the apex and feebly converging and very slightly sinuate to the base; sur- face with a broad and shallow, densely sculptured channel inclosing a small central callus and ending at base in a broad deep cavity, thence to the sides coarsely, very densely punctate, with about three callous spots, obliquely impressed behind the middle; elytra not evidently wider than the prothorax, only slightly more than twice as long as wide, gradually rounded and converging at the sides in apical two-fifths, the prolongation gradually formed but rather broad and short, the cleft narrow and deep; surface somewhat coarsely subsulcate almost through- out the width, the sulci rather coarsely and irregularly, not distinctly punctate, the elevations numerous and strong but for most part small and irregular, the depressed spots brighter metallic, strongly punctate; prosternum ( 9) as broadly concave and roughly sculptured as in the STUDIES IN THE AMERICAN BUPRESTIDif: I5I male of other species, the concavity limited only by a narrow ridge at each side, the median apical tooth triangular but fmely aciculate. Length (9) 17.5 mm.; width 6.2 mm. Lake Superior (Whitefish Point) rigida n. sp. Form very stout, moderately convex, the lustre feeble, deep black throughout above, feebly metallic at the side margins, bronzed beneath; head coarsely, densely and subevenly punctato-rugose, the eyes moderately developed but only slightly convex; prothorax in outline nearly as in rigida but with the lateral inflation at the middle, the sulcus not quite so broad but ending posteriorly in a similar large transverse cavity, the central callus prolonged posteriorly in a fine line; surface from the bounding ridges to the sides coarsely, very densely and confusedly but evenly punctato-rugose, without callous spots and not obliquely im- pressed behind the middle; elytra at least a third wider than the pro- thorax, scarcely more than twice as long as wide, the sides rather ab- ruptly rounding and strongly converging posteriorly in more than apical two-fifths or from only slightly behind the middle, the prolongation gradually formed, not very long and somewhat finely acuminate; sur- face nearly even, with fine, unevenly punctured sulci throughout the width, densely rugulose laterally, the elevations small, few in number and inconspicuous; under surface ( 9 ) v^dth the prosternal concavity not quite so broad as in rigida and more coarsely sculptured, the ridges of the fifth ventral uniting at the apex, the latter narrow, wholly occupied by the shallow sinus, the lateral lobes acute, the median tooth narrowly triangular, very acute and extending beyond the lateral lobes. Length ( 9) ig.o mm.; width 6.8 mm. Ontario (Port Hope), — C. J. S. Beth- une) tetrica n. sp. I have identified as prolongata Lee, a female specimen from an unrecorded locality in British America. It answers the description very w^ell, except as to the proportions of the prothorax, the latter being evidently less than twice as wide as long, but this is unimpor- tant, since the width in terms of length was habitually overdrawn by LeConte, as in the case for example of Spinthoptera valens, pre- viously referred to. The elytra are somew^hat inflated behind the middle, the sides gradually rounding and rather rapidly converging in more than apical two-fifths, the attenuated apex much prolonged and rather acuminate but gradually formed and very different from the same part in divaricata and allies, w^ith vv^hich the species was united in complete synonymy by Kerremans. The above description of prolongata is wholly drawn from the characterization published by LeConte in his monograph (1859). If we admit that tenehrica Kirby, belongs to this group, and not near tenehrosa, of the next group, and the few published characters seem to warrant this assumption, it 152 CASEY must be closely allied to polongaia though probably a smaller species. The statement of Kirby that the elevated spots are only evident near the sides of the elytra, refers to the greater prominence observable, as a rule, in the submarginal interval behind the middle; this is par- ticularly pronounced in certain species of Buprestis, as well, and is a general character, at least in this part of the family. Longipennis bears much the same relationship to prolongata, that pertiimx does to subcuprea, and in both cases the subspecific type comes from the same locality as the species to which it is attached. In the latter case, the long and parallel-winged type of the assumably subordinate form, is a male, having besides some striking differences in the prosternal groove, so that I have given it subspecific rank provisionally. In the case of longipennis, however, the aberrant long-winged form is represented by both male and female. It may be that we have here evidences of true asexual dimorphism, but the material at hand is not sufficient to decide this at present. The general appearance of the subspecies in each case departs remarkably from that of the species, to a greater degree in fact than the general differences in facies distinguishing a number of allied though evidently distinct species. This is appar- ently therefore one of those puzzles that occasionally arise in taxo- nomic work, with incomplete material and in the absence of biologic evidence. Group IV — Type horni Cr. This is a small and very local group, peculiar to the Southern Paci- fic coast fauna. The only described species is the following: — Form moderately stout, elongate-oval, not very convex or shining, obscure bronze above, with the margins of the pronotum and elytra and the occiput behind the eyes generally brighter or greenish, the under surface brighter cupreous; head coarsely punctato-rugose, strongly impressed along the median line throughout, rather more hair)^ in the male though sparsely, the eyes well developed and rather prominent; prothorax not quite twice as wide as long, the sides parallel and straight to the middle, there rounding and thence strongly converging to the apex, which is feebly sinuate; surface moderately impressed and more punctured along the median line, the impression with a feeble slender callus, coarsely, deeply and densely punctate with intermingled irregular callous spots from the broad and feebly tumescent ridges to each side and very broadly, diffusely impressed obliquely behind the middle, the basal pit deep, single; elytra barely at all intlated behind, twice as long as wide and only slightly wider than the prothorax, the sides grad- STUDIES IN THE AMERICAN BUPRESTID^E 1 53 ually rounding and converging in apica! two-fifths, the short and broad prolongation rather abruptly formed, more cupreous and with the outer angles moderately rounded, the sutural feebly dentiform; surface sparsely pubescent, with moderate shallow, closely and coarsely punc- tate sulci in about inner half, densely, coarsely punctured, confluently so laterally, the callous spots strong but small and very sparse; under surface rugose, deeply sulcate along the axial line including the first ventral, the fifth with two vestigial ridges, the apex transversely emar- ginate (c?), or broadly and obtusely rounded (9). Length 16.5-18.7 mm. ; width 6.0-6.8 mm. California (Tulare Co.) horni Cr. A — Similar but larger, stouter, more dusky and obscure in coloration, the head more coarsely rugose, the prothorax feebly inflated near the middle of the sides, the medial sulcus obsolescent, the punctures laterally denser and the anterior callus near lateral fourth much stronger; elytra nearly similar but with the outer angles of the pro- longation much more broadly rounded; fifth ventral (9) without trace of ridges and with the apex more acuminate and very narrowly rounded. Length (9) 17.7 mm.; width 6.g mm. California (Ka- weah) ampliata n. subsp. The variety, mentioned by Crotch, having the last ventral segment of the female tridentate at tip, I have not seen. It is probably a widely distinct species if the sexual characters are correctly stated. Group V — Type tenebrosa Kirby. The general habitus in this group dififers greatly from that charac- izing any of the preceding, the body being shorter and relatively broader, with the prothorax more transverse as a rule, generally in- flated at the sides before the middle and with the surface much more unevenly impressed and sulcate. The elytra are more or less rapidly narrowed behind and briefly produced, the prolongation either nar- row and acuminate as in tenebrosa and related species, or more broadly obtuse as in crassicollis, where it approaches the form described above in horni, but the apices are in all instances entire at tip. The species are rather numerous and more fully described by various early authors than in either the obscura or divaricata groups. After rather atten- tive study of the literature and of such material as is at hand, I would propose the following arrangement of those discovered thus far: — Pronotal sulcus generally broader, densely punctate, having a small central callus by which it is in no measure interrupted ; middle tibiae of the male unmodified ; head with an elevated transverse line between the eyes 2 Pronotal sulcus narrower, with a large central callus by which its continuity 154 CASEY is more or less completely interrupted; middle tibiae of the male ob- tusely dentate within; head without a callous band between the eyes 7 2 — Prothorax subparallel, only feebly inflated at the sides near apical third. Body elongate, subparallel, moderately convex, dull in lustre, dark brown or blackish and without evident metallic reflection above, similar beneath though more shining, nearly glabrous above, the head densely but not very coarsely punctato-rugose, with a strong callous transverse band between the eyes and two longitudinal callous areas at base ; eyes moderate, not very prominent; prothorax three-fifths wider than long, the sides parallel and very feebly sinuate from the anterior inflation to the base, converging anteriorly ; surf ace finely, densely punctate through- out, except the rather approximate medial ridges, and thence to the sides obhquely, though very moderately, impressed behind the middle and with an elongate anterior and small basal callus in a line nearer the medial ridges than the outer margin; central callus small and finely prolonged posteriorly; elytra distinctly wider than the prothorax, twice as long as wide, the sides parallel, gradually rounding and oblique in posterior two-fifths, becoming only very faintly sinuate near the obtusely acuminate apices, which are not evidently prolonged; surface finely, deeply, very closely and uniformly punctato-cribrate throughout, with series traceable to the sides of very coarse and well spaced per- forate punctures, also having some elongate callous smooth lines on the alternate intervals, especially toward the suture; under surface not coarsely ])unctato-rugulose laterally, smooth and but feebly impressed axially, the male with the middle tibiae unmodified, the apex of the fifth ventral with a shallow transverse emargination, the female with the apex narrowly and prominently rounded and entire. Length 1 1.0-13.7 mm.; width 4.0-4.8 mm. Massachusetts, Pennsylvania and District of Columbia. [==Bupresiis transversa Say] punctulata Schonh, A — Similar to the above but larger and stouter, the head larger, with relatively smaller eyes, the transverse ridge finer and more acutely elevated but not quite so strong, the prothorax nearly similar but with the sublateral line of callus only narrowly interrupted or entire; sides of the elytra similarly obUque posteriorly but with thesinuation rather more obvious; prosternum of the male broadly, feebly con- cave and closely, coarsely punctured, scarcely at all hairy, the abdo- men strongly punctured throughout. Length (c?) 14.8 mm.; width 5.6 mm. North Carolina (Southern Pines) pinorum n. subsp. Prothorax constricted and sinuate at the sides basally, broadly inflated before the middle 3 3 — Elytral apices rather broad and obtuse, not evidently prolonged, the form being nearly as in the preceding species. Body stout, o' long, moderately convex, feebly shining, dark cupreous-brown above, some- times with small green areas and occasionally with the elytra almost entirely green, the under surface dark purplish-cupreous; head coarsely, densely punctato-rugose, with a sharply elevated transverse biangulate callous band between the eyes and two small approximate callous spots at base; eyes rather prominent; prothorax three-fourths wider than long, STUDIES IN THE AMERICAN BUPRESTID^. 155 very densely, uniformly punctured, with two strong median and two sublateral callous vitt;e, the latter broadly interrupted by the concave impressions, also with a small central callus and an irregular elevation near each side margin; scutellum small, strongly transverse; elytra much viider than the prothorax, three-fourths longer than wide, the sides rather abruptly rounding and strongly oblique in apical third, becoming evidently sinuate near the tips, the latter transversely trun- cate, with the angles not at all rounded, the sutural slightly prominent; surface extremely densely, not coarsely punctate throughout, with unimpressed series of rather large but more or less obscure punctures though coarse and perforate in certain parts near the suture, the alter- nate intervals with moderately elevated and smooth elongate-oblong callous spots, varying greatly in size; under surface ( 9 ) rather coarsely and subrugosely but not very densely punctate, the axial line only feebly impressed, the prosternum broadly, very feebly concave and coarsely, densely punctate throughout, the apex of the fifth ventral narrowly rounded, with two small deep approximate incisures, the me- dian tooth parallel,^truncate; male not at hand. Length 15. 5-16. 5 mm.; width 6.0-6.6 mm. Oregon and Washington State — eight specimens. [ = calif ornica Cr., i. litt .] crassicollis Lee. A — Body larger, similar in general form, sculpture and coloration, the head relatively a little smaller, with less prominent eyes, the sur- face more confusedly and strongly rugose, the sides of the elytra more gradually rounding and less strongly oblique in apical two- fifths, the punctured series less distinct; under surface nearly similar; legs stout. Length (9) 17.5 mm.; width 6.9 mm. Washington State hesperica n. subsp. Elytral apices more gradually and finely attenuate, very narrow at tip and more or less evidently though but slightly prolonged 4 4 — Upper surface black, without metallic reflection, at least in morio 5 Upper surface dark brown, with feeble subcupreous or aeneous lustre, the under surface rather bright cupreous 6 5 — Body very much as in divaricata (!) but much smaller, the upper surface confluently punctured, with only the elevated parts glossy, the lower surface bronzed-copper and glossy; mouth and antenna; bronzed, the eyes black; front sculptured as in divaricata; prothorax uneven, with shallow impressions and a broad dorsal channel, distinctly bisinuate at base; surface with levigated elevations; scutellum very minute, im- pressed; elytra divaricate, suddenly attenuated at the apices, which are rounded and each with a single ridge; surface rough, with many con- catenated and levigated irregular elevations, the sides of the apices bronzed; breast channeled, the first abdominal segment less conspicu- ously; prosternum nearly an isosceles triangle; abdominal apex with three short teeth. Length 15.5 mm. British America (latitude 65° and in the Rocky Mts.). [ =Stenuris tenehrosa Kirby]. tenebrosa Kirby Body oblong, rather stout, moderately convex, dull, deep black throughout 156 CASEY above, deep metallic magenta with feeble medial aeneous reflection beneath; head not very coarsely, densely and unevenly punctato-rugose, with an uneven transverse callous band between the eyes and two un- even basal callous areas, the eyes moderately prominent, pale in the type; prothorax four- fifths wider than long, the sides strongly and sub- evenly arcuate, becoming about parallel for only a very short distance at base, widest slightly before the middle, the sulcus broadly concave, closely punctate and with a small central callus, the surface from the ridges to the sides very uneven, densely punctate and rugose, strongly impressed behind the middle, the exterior callous line broadly interrupted, the lateral ridge from the basal angles to the middle strong; scutellum whitish and coria- ceous in the type, transverse, semicircularly rounded behind ; elytra barely at all wider than the prothorax, four-fifthslongerthanwide, parallel, the sides gradually rounding and feebly oblique in posterior two-fifths, becoming scarcely at all sinuate toward the narrow apices, which are truncate with rounded angles, their discal carina not extending to the tip; surface not coarsely, very densely punctured laterally, sparsely so and in clusters suturally, with even series throughout the width of very large perforate punctures, which are not at all connected in the series by any kind of a stria; elevated spots few in number, small and with very irregular outline; under surface (c?) strongly punctato- rugulose, the presternum broad and flat, very densely, coarsely punc- tured, the metasternum broadly concave and also densely punctured, the first ventral simply broadly flattened medially, the middle tibiae unmodified, the fifth ventral with a shallow, broadly sinuate apical emargination, the angles acute. Length 15.0 mm.; width 5.5 mm. Lake Superior (Whitefish Point) morio n. sp. 6 — Form oblong, attenuate behind, moderately convex, not very stout, being somewhat as in ptmctulata in general form and sculpture; head densely and confusedly punctate, with a very irregular band of low callus be- tween the eyes, the latter rather small and scarcely at all prominent; prothorax three-fourths wider than long, the sides evenly rounded and inflated anteriorly, becoming nearly straight and parallel in almost basal third; surface with the median ridges very narrow, strong, the inclosed sulcus deep, punctate, with a smaU central callus, the surface thence laterally very densely and confusedly but shallowly punctate, deeply impressed behind the middle interrupting the narrow line of caUus, the submarginal ridge evident but very irregular ; scutellum black, dull, transverse; elytra quite obviously wider than the prothorax, twice as long as wide, the parallel sides rather rapidly rounding and thence moderately oblique, becoming very broadly but distinctly sinuate to the narrowly and evidently prolonged apices, which are narrow and rounded at tip; surface punctured nearly as in morio and with series of perforate coarse punctures, which are however smaller and more compressed and only very obvious suturally, the raised spots more numerous and very irregular, mostly elongate and coarsely crenated at their sides by the punctures; under surface (?) rugulosely punc- tate, the densely punctate presternum flat, the fifth ventral gradually and acutely attenuate, with the apex very narrowly, subangularly STUDIES IN THE AMERICAN BUPRESTID^ 157 rounded and entire; male not at hand; Length 14.4 mm.; width 5.0 mm. Michigan (at Lake Superior). [=lacustris Lee. (fide Crotch)]. lugubris Lee. A — Body nearly as in lugubris but more prolonged and more gradually attenuate posteriorly, the head nearly similar, the prothorax shorter and more transverse, nearly twice as wide as long, almost similar in outline and sculpture but with the two median ridges less elevated, the inclosed sulcus shallower, the central callus prolonged in a fine callous line to the base; elytra with the elongate callous spots more numerous, very irregularly crenulated along their sides by the coarse punctures, which are even more conspicuous than in lugubris but less perforate than in fnorio, the sides posteriorly very gradually converging to the narrow rounded apices but becoming only scarcely visibly sinuate; prosternum (c?) broad, just visibly concave and very densely punctate though scarsely hairy throughout; middle tibiae unmodified, the fifth ventral with a shallow transverse emargination. Length (6^) 15.0 mm.; width 5.25 mm. Michigan (Whitefish Point, Lake Superior) austera n. subsp. Form and size nearly as in punctulata, brassy-brown above; head dull cupreous, densely rugulose, slightly indented on the front and with a transverse interrupted polished callous line between the eyes; antennae obtusely serrate, rich cupreous; prothorax transverse, slightly emargi- nate at apex, feebly bisinuate at base, where it is narrower than the base of the elytra, the posterior angles acute, the sides straight near the base, strongly dilated beyond the middle, dull cupreous, densely punc- tured, with four longitudinal polished black lines, the middle two entire, the outer interrupted before the middle or before an oblique fossula and appearing slightly again at base; there is also a narrow polished marginal line from the tip of the hind angles to near the middle of the lateral dilatation; scutellum transverse, black; elytra brassy-brown or black, densely punctate, with series of large remote punctures and many irregular linear black polished elevations; tip short, entire, the legs and under surface cupreous, shining, rugosely punctured, the post-pectus canaliculate. Length 14.0 mm. ; width 6.0 mm. Virginia. [ =consobrina Mels.] dumolini Lap. -Gory Form nearly similar, asneous; prothorax twice as wide as long, strongly dilated at the sides, which are rounded anteriorly, sinuate posteriorly, roughly punctate, unequal, with a short carina at the posterior angles, callous spots at apex and base and two nearly smooth costos, the dorsal sulcus broad and deep, with a small central callus; elytra punctured, the internal stris coarsely punctate, the alternate intervals with ele- vated shining spaces, many of which are connected by elevated trans- verse rugae, the tips briefly prolonged, rounded; under surface bright copper}', the prosternum flat, densely punctured, the metasternum feebly sulcate, and, like the unimpressed first ventral, sparsely punctate; male with the last ventral truncato-emarginate at apex, the middle tibiae simple; female with the fifth ventral rounded at tip. Length 13. 0-16.0 mm. Maine and New York. [=/w&erc2- press both erecta and debilis, but I do not hold this to be justifiable; the former is the widely dififused eastern form, with brilliantly cup- reous apex of the elytra and blue tarsi, usually identified in cabinets as cyanipes, which however is a very different and far western species, probably rather rare in collections. Anataxis n. gen. Although the prosternum in this genus is almost exactly as mPoecilo- nota, — broadly convex between the narrow lateral sulci, — both the meta- sternum and first ventral are broadly and evenly convex, and, that we have here a remarkably isolated type, not at all closely related to any- thing that precedes, is amply shown in addition by the transversely parallel hind coxae, parallel inner margins of the eyes and by the antennal structure. The body is cuneiform, convex and not unlike AcmcBodera in general outline, the evenly convex front feebly impressed medially, the epistoma depressed and sinuato-truncate, the labrum short, sinuate and closely, stiffly pubescent, the mentum transverse, corneous, metallic and medially produced in a narrowly rounded prominence at apex and the maxillary palpi slender and moderate in length. The antennae are short, compact, somewhat as in Dicerca but with the third joint longer than the second or fourth, the outer joints transversely and obtusely serriform, their outer sides smooth but having each a very large oval sensory intero-apical fossa, and another, smaller and transversely oval, near the base of the joint, with a number of coarse sensitive punctures between; they are inserted in small and widely separated foveas, which have their upper margins slightly j)romincnt. The prothorax is evenly convex, excepting a feeble STUDIES IN THE AMERICAN BUPRESTID^ 17I impression along the middle near the base, the scutellum small, slightly transverse, flat and rounded behind, the elytra cribrate and with some fme and only slightly apparent, somewhat more convex punctureless hnes, the sides posteriorly serrulate, the tips not at all prolonged, rather narrowly rounding to the suture and serrulate. The legs are short and slender, the hind tarsi short, slender, with the basal joint but slightly elongate and the first ventral suture is straight and very fine, almost obliterated in the middle. It is of course within the most evident possibility, that in placing Anataxis here, the genus may be unwarrantably removed from some of the preceding genera, with which it may in reality and despite its very different appearance be more closely related. This surmise occurs on considering the actual outline of the compact outer antennal joints, the latter having a feebly defined trvmcature of the lower edge, which posteriorly thereto converges much more rapidly to the base, in conjunction with the complex system of sensory fossae, which might suggest that it be placed in closer proximity to Psiloptera. In considering the figure of Anthaxia gigas, the type of AgcBocera Wat., as given by Laporte and Gory, it is observed to differ in several external characters, such as the four pronounced costae of each elytron, the triple longitudinal impression of the pronotum and the apparent absence of serrulation of the postero-lateral margins of the elytra. Making due allowance for inaccuracy of drawing, it seems to me prob- able that these marked external incongruities betoken other differ- ences in the organs of the head and structure of the under surface, which I am unable to express at present, but which may be even more significant. The single species of Anataxis may be described as fol- lows : — Body moderately slender, convex, rather shining, bright green in color throughout, the under surface more polished and partially cuprascent, glabrous; head rather finely, closely and evenly punctate, the eyes well developed but not prominent, the antennae but little longer than the head, black with feeble greenish lustre; prothorax two-thirds wider than long, the sides broadly, evenly rounding and converging ante- riorly, becoming very gradually parallel basally, the base broadly bisin- uate, the apex very obsoletely bisinuately truncate; surface rather finely but strongly, loosely punctate, gradually more coarsely, densely so laterally, even, the feeble impression behind the centre ending, at some distance before the middle of the base, in a small punctiform fovea, the bottom of the impression finely striiform near the puncture; 172 CASEY elytra at base as wide as the thoracic base to somewhat narrower, be- tween three and four times as long as the prothorax, the sides straight and scarcely visibly converging to apical two-fifths, there very grad- ually rounding and arcuately converging to the very narrow stronger arcuation extending to the sutural angles; surface irregularly but sub- equally cribrate, with rather small and deep punctures, nearly wanting in two broader, feebly convex Unes on each near the suture, dense laterally, the submarginal prominent ridge rather evident; under sur- face finely, more feebly punctate, rather closely so laterally, the surface smooth and convex, the fifth ventral transversely truncate at the rather narrow apex, with the ends of the truncature minutely dentiform, the edge impunctate to a fine transverse striiform parallel line a short dis- tance therefrom, apparently not differing noticeably in the two sexes. Length 9.0-12.2 mm. ; width 3.0-4.3 mm. Texas (El Paso). [=Ha- lecia gentilis Horn and Agceocera gentilis Waterh.] gentilis Horn The male is smaller and narrower than the female, if I have prop- erly identified the sexes, but does not differ otherwise, except in having the prothorax more parallel and less narrowed anteriorly. The single male before me has a small irregular depression at each side of the median line near anterior fourth, but these impressions are probably adventitious. Nanularia n. gen. This genus and the two following form a group, altogether distinct from any of those that precede in having the mesosternum separated from the metasternum by a deep cleft; there is, however, little or no harmony between these three genera in other structural characters, and I am unable at present to assign a definite value to the character as a means of dividing the entire family into lesser subdivisions. The body in Nanularia is small, convex and deep, not unlike Anataxis in this respect, as well as in sculpture, except that the outline is par- allel and not cuneate. The head is relatively large and but little narrower than the prothorax, the front even, but with a narrowly impressed and subentire median line, the eyes moderate, with their inner margins converging upward though very feebly, the epistoma not depressed, broadly, feebly sinuate, the labrum short, subcoria- ceous, feebly sinuate, sparsely punctate and with short stiff sparse hairs, the mentum short, very transverse and very broadly, parabolic- ally rounded, the palpi rather slender. The antennae are inserted in small and widely separated fovea:, each margined above by a small inconspicuous oblique ridge, the ridges tending however to join each STUDIES IN THE AMERICAN BUPRESTID^ I73 Other transversely, especially in the male; in the female there is only a small median part of the transverse ridge visible; they are short and compact but not very stout, the third joint elongate, the outer joints in- cluding the fourth, serriform but unusually convex and not much com- pressed, with their lower margins obliquely though somewhat vaguely and narrowly truncate, then very rapidly converging to the base, their inferior surface entirely opaque from very dense punctuation and each with a distinct circular sensory fossa on the oblique apical sur- face, though more visible from within; the last joint in both sexes is strongly and narrowly prolonged on the inner side, forming a narrow process nearly as in Gyascutus. The pronotum is almost even, the scutellum small but distinct, rounded and medially foveate, the elytra cribrate, imperfectly and very finely striate, not distinctly serrate at the sides and with the apices not at all prolonged and narrowly trun- cate. The under surface is strongly convex, the prosternum narrow, feebly convex, not sulcate at each side and densely sculptured through- out, the first ventral suture perfectly rectilinear, fine but rather dis- tinct, the first segment not free, the last narrowly and evenly rounded at tip, more broadly in the male. The legs are short and slender, the hind tarsi three-fourths as long as the tibiae, slender, with the basal joint but slightly elongate, although nearly as long as the next two combined. The single type known to me may be described as fol- lows : — Body parallel, subcylindric, scarcely shining, obscure cupreous to blackish- brown, the female subglabrous, the male with short though rather numerous hairs beneath and on the head, the latter evenly and deeply, rather densely cribrate, more rugose in the male, the antennae about a third longer than the head; prothorax a third wider than long, sub- parallel, the sides feebly rounded anteriorly, becoming very broadly, feebly sinuate posteriorly, the apex but little narrower than the base, broadly and evenly arcuato-truncate, the base broadly, transversely lobed medially, sinuate laterally, the surface gradually very deeply declivious laterally, somewhat coarsely, deeply and closely cribrate throughout, very feebly impressed subapically along the median line, the entire posterior dorsal part of the surface very feebly impressed, so that the margin at the median lobe is turned slightly upward and is less punctate, sometimes also having a small sinus at the scutellum; elytra but barely wider than the prothorax and about three times as long, the sides parallel to apical third, there broadly rounding and gradually arcuately converging to the apical truncature, which is wider in the female; surface subevenly and closely cribrate, the punctures not as large as those of the pronotum, also having fine and unimpressed 174 CASEY striae arranged in pairs and more or less evident throughout the width ; under surface rather finely but deeply, closely cribrate, the abdomen very finely punctate, sloping upward posteriorly. Length 7.9-9.8 mm. ; width 2.5-3.1 mm. California (San Diego Co., — Poway). cupreofusca n. sp. The following species is provisionally included : — Form oblong, subcylindric, gradually narrowed posteriorly, color bronze with slight aeneo-cupreous lustre; front slightly convex, densely and coarsely punctured; epistoma broadly emarginate, the antennal ridges short; antennae shorter than the head and prothorax, serrate, the ter- minal joint obtuse at tip; prothorax slightly broader than long, cylin- dric, the sides slightly arcuate in front, feebly sinuate at basal half, which is subacutely margined, the hind angles subacute posteriorly; apex truncate, the base bisinuate and at middle subtruncate; surface moderately convex, coarsely and deeply but not densely punctate; scutellum semicircular, nearly smooth; elytra subcylindric, parallel, gradually narrowed at apical third, the margin near the apex finely serrate, the apex obtuse; surface densely punctate and with traces of striae at the sides and apex; under surface coarsely but not densely punctured and with a few short cinereous hairs; posterior tarsi with the first joint feebly compressed and not greatly longer than the second. Length 9.0-12.0 mm. California (San Joaquin Valley). [=Gyas- cutus californicus Horn] californica Horn Among the remarks following the description of californica, Dr. Horn states, in referring to the species of Gyascutus, that " traces of a false joint at the end of the eleventh occur in the antennae, excepting in californicus and sphenicus, in which the eleventh joint is simply ob- tuse as in Psiloptera. " The form of the eleventh joint in californica must therefore depart very radically from that characterizing cupreo- fusca, and the allusion to an obtuse eleventh joint in sphenicus {Hip- pomelas) refers, as I have stated under that genas, to the male alone. Cinyra Lap.-Gory. The type of this genus is stated, in the preceding table of genera, to be the Brazilian Buprestis elongata, of Laporte and Gory, but the diag- nosis there given is drawn from our small aberrant Dicerca gracilipes, of Melsheimer, or more accurately from a very similar species described below under the name macilenta, and the corresponding characters of elongata are unknown to me. So there is a possibility, which amounts almost to a certainty, when we consider the very different style of sculpture of gracilipes, that the latter may constitute a different genus from the true Cinyra, where it was first placed by LeConte; STUDIES IN THE AMERICAN BUPRESTID.F, 175 in the Kerremans Catalogue it appears under tiic genus Halecia. I will therefore assign our species to the following subgenus of Cinyra: — Body slender, the elytra roughly and unevenly sculptured, having large, indefinitely limited and feebly depressed arcolie of finer closer punctu- ation, the sides posteriorly without trace of serrulation and perfectly even; posterior tarsi with the pads of the first three joints small and depending vertically from the under part of the extreme apices of the joints, the fourth gradually oblique and lamellate under the fifth, impressed along the median part of the surface and evenly sinuato- truncate at tip. [Type Dicerca gracilipes Mels.j. Spectralia n. subgen. The type is rather rare and I have in my collection but a single example of the subgenus, which however comes so far from satisfying the original diagnosis of Melsheimer, that it seems necessary to regard it as a different though closely allied species, of which the description appended below is made unusually complete, in order to bring out possible generic as well as specific characters. The head is as wide as the thoracic apex, vertical, the front flat, the epistoma not depressed, deeply and parabolically sinuate at apex, the labrum corneous, sinuate and with an abrupt pale coriaceous margin at apex, the mentum short, wholly corneous and roughly sculptured, truncate and trapezoidal, the palpi slender, the eyes with their inner margins evidently converg- ing upward. The antennae are somewhat as in Buprestis though still more slender and elongate, inserted in small and widely separated foramina, which are concealed from above by the sharply defined and somewhat prominent though porrect frontal margins above them. The prosternum is moderately wide, flat or very feebly convex, ha\nng throughout a very fine marginal bead; it is abruptly and obliquely narrowed posteriorly to a rather narrow apex, which does not quite attain the rounded apex of the metasternum in the wide cleft between the lobes of the mesosternum.' The first ventral segment is fully as long as the next two combined, with the suture fine but distinct and rectilinear. The descriptions follow: — Form slender, subcylindric, dark brassy-brown above; head bright cupreous, roughly chagrined, with a transverse arcuate glossy line between the eyes, which are black-brown; prothorax transverse, with the sides rectilinear, the apex truncate and the base bisinuate; surface with an obvious entire furrow, rugosely punctate; scutellum obtuse, triangular; elytra faintly striate, with about four or five longitudinal raised lines, the interstices rugose, punctulate; surface with a distinct cupreous impressed spot a little before the middle and an obsolete double one Proc. Wash. Acad. Sci., April, 1909. 176 CASEY behind the middle; sides parallel from the base to the apical curve, the sides posteriorly not serrate, the tips much narrowed, feebly bidentate; feet and under surface bright cupreous, rugosely punctate; tibiae long, slender; venter almost rounded at tip; pectus feebly canaUculate. Length 10. o mm.; width 3.5 mm. Pennsylvania. [ =Dicerca gracilipes Mels.] gracilipes Mels. Form slender, moderately convex, fusoid, above slightly shining, blackish with feeble cupreous reflections, except the elytral foveolae which are greenish, beneath more cupreous and less glabrous, each puncture hav- ing a small decumbent hair; head nearly as wide as the thoracic apex, obscure in color, the front flat, rather finely, densely, almost uniformly punctato-rugulose and evidently pubescent, without trace of transverse callus but with the median line shghtly elevated and smooth near the centre; antennae piceous, with slight cupreous lustre except apically, very slender, extending somewhat beyond the thoracic base; eyes well developed, rather convex and prominent; epistoma cupreous, the sinus small, deep, parabolic; prothorax widest at base, nearly one-half wider than long, feebly trapezoidal, the sides very feebly arcuate, becoming straight basally, the angles not at all everted; apex broadly sinuate, the base arcuate, becoming feebly sinuate near the sides; surface strongly, rather sparsely and subevenly punctate medially, with a deep entire median sulcus which is broadly impressed posteriorly, becoming more acute anteriorly, and another, near each side, very feeble anteriorly but becoming deep at base, concave and densely punctate and bounded externally by a tumid and less punctured ridge, which is obsolescent anteriorly; scutellum small, flat, opaque, blackish; elytra but Httle wider than the prothorax and about four times as long, separated along the inner sides in the type from basal fifth to the apex; sides par- allel and nearly straight for three-fifths, then very gradually rounded and convergent, even and without trace of serrulation to the narrowly transverse sinuato-truncate apices, which are feebly bispiculose; sur- face with rather coarse, uneven interrupted striae, except laterally, with the narrow intervals alternately a little more prominent, strongly so apically, very unevenly but closely, rather coarsely punctate and rugu- lose, more sparsely and coarsely on the general surface but more densely and finely in the shallow and very uneven vague depressions, of which there is one very large and obliquely transverse before the middle and some smaller from the middle to the apex, the small punc- tate patches well separated along the smoother sutural intervals much as in Dicerca; prosternum perfectly flat, transversely truncate ante- riorly, rather coarsely, closely and evenly punctured throughout; entire metastcrnum equally densely and uniformly but less coarsely punctate, the abdomen much more sparsely, especially toward the middle, and more polished and ratlicr narrowly sinuato-truncate ajncally in the type; legs long, very slender, the basal joint of the hind tarsi as long as the last three combined. Length 10.5 mm.; width 3.2 mm. Texas. macilenta n. sp. The description of gracilipes is drawn directly from the originally STUDIES IN THE AMERICAN BUPRESTID.E 177 published characters. It is stated by LeConte that the anterior tibiae of the male are serrate within; in the type of macilenta the anterior tibiai are almost straight, with the inner margin very finely spiculo- serrate throughout the length; the middle tibiae are longer, feebly arcuate and also finely and closely but less evidently spiculose through- out the length. The hind coxae are rapidly dilated inwardly as usual. Mr. Chas. Schaeffer has lately (Journ. N. Y. Ent. Soc, 1904, p. 205) described a Cinyra prosternalis from Texas, and, still more recently, Cinyra purpurascens, from Lower California (Bull. Br. Inst., I, p. 127, 1905). Without being able to decide upon the proper sys- tematic position of these species, it may be safely stated that neither of them can be assigned to Spectralia, the first differing in its feebly and apparently more broadly sinuate epistoma, obsolescent thoracic sul- cus and very different elytral sculpture, and the latter because of its very short elytra, absence of thoracic sulcus, shorter hind tarsi, dif- ferently sculptured abdomen and many other important characters. The antennae in neither of these species is so described as to afford a clue as to whether they belong in this vicinity, or are more closely allied to Anataxis or Nanularia. It is probable that they are not properly assignable to Cinyra, even in its broadest sense. Trachykele Mars. The species of this genus recently formed the subject of a review by Mr. H. C. Fall (Ent. News, 1906, p. 160), in which the characters of the four known species were given in considerable detail, but no mention was made of the generic characters, nor was any opinion advanced as to the relationships of the genus or its proper position in the family. The separation of the meso- and metasterna by a cleft- like suture, is apparently quite unusual in this family and present only in Nanularia, Cinyra and Trachykele, of those genera discussed in the present paper, but, because of my insuflScient knowledge of exotic types, I am unable to put forward any useful suggestions as to the relative significance of this character in a general analysis of the family. In any event, however, it clearly isolates these three genera from those above them in the table given at the beginning of this paper, and there seems to be, besides, a pronounced tendency in them toward a more mobile connection between the first and second abdominal segments, this feature being especially pronounced in Trachykele, lyS CASEY where the first segment appears to be virtually free, the first suture not differing to external view from the others. The species are extremely rare in collections but, in greater likeli- hood than actual rarity, perhaps, they may be abundant only at certain very short periods of the year or in relatively few limited local- ities, like Buprestis conjluenta and some others of this family, and as also frequently noted in the Cerambycidae. The genus was conse- quently wholly unknown to me in nature, until Prof. Wickham very kindly lent me the only example that he had secured during many years of skillful collecting. This specimen has served as the type of the diagnosis given in the table of genera on p. 53; it appears to satisfy the general characters of opulenta Fall, which is closely allied to hlon- deli Mars., the true type of Trachykele, and was taken at Nanaimo, Vancouver Island. It is glabrous, very densely and rather finely cribrate and bright green throughout, the elytra with a few small and irregular black spots. These spots are observed under higher power of amplification to be due, not to pubescence, but to actual black- ness of the integument, accompanied by a strong granulato-reticula- tion of the surface, which gives them their velvety appearance. PROCEEDINGS OF THE WASHINGTON ACADEMY OF SCIENCES Vol. XI, No. 3, pp. 179-238 August 14, 1909* THE STRATIGRAPHIC RELATIONS AND PALEONTOL- OGY OF THE "HELL CREEK BEDS," "CERATOPS BEDS" AND EQUIVALENTS, AND THEIR REFERENCE TO THE FORT UNION FORMATION.i By F. H. Knowlton. Contents. Introduction 180 Problem and conclusion 180 Historical summary 181 Areal distribution and lithologic character of the Fort Union formation . . 182 Division of the Fort Union into two members 183 Areal distribution and paleontologic contents of the lower member of the Fort Union 184 1. Hell Creek, Montana 184 2. Miles City, Montana, and vicinity 186 3. Forsyth, Montana 190 4. Custer, Montana, and vicinity 192 5. Bull Mountain, Montana, area 192 6. Melville, Montana, area 192 7. Red Lodge, Montana, area. 195 8. Glendive, Montana, and vicinity 196 *Author's separates of this paper were distributed June 25, 1909. ^ Pubhshed with the permission of the Director of the U. S. Geological Survey. This paper was read before the National Academy of Sciences on April 22, 1909, under the title; " Stratigraphic relations and Paleontology of the lower member of the Fort Union Formation, " the title finally selected having been decided upon too late for printing in the Programme of the Academy meetings. Proc. Wash. Acad. Sci., August, rgog. l8o KNOWLTON 9. Glendive, Montana, to Medora, North Dakota 199 10. Bismarck, North Dakota, and vicinity 200 11. Yule, Billings County, North Dakota, and vicinity 201 12. Probable area in Northwestern South Dakota 202 13. Weston County, Wyoming, area 203 14. Northward extension of AVeston County area to connect with Miles City area 2 04 15. Converse County, Wyoming, area 205 16. Area to the eastward of the Bighorn Mountains, Wyoming 209 1 7 . Bighorn Basin, Wyoming 212 18. Probable distribution of lower Fort Union in other areas 215 Relations between the lower member of the Fort Union and underlying formations 216 Relations between lower and upper members of Fort Union 217 Stratigraphic relations of the lower member of the Fort Union to the Laramie 217 Paleontological characters of the lower Fort Union: 218 1. Plants 218 a. Plants identified in lower Fort Union 219 b. Paleobotanical proof of the Eocene age of the Fort Union flora 224 2. Invertebrate evidence 226 3. Vertebrate evidence 229 a. Dinosaurs 229 b. Mammals 233 c. Chelonians 235 d. Fishes 236 Line between Cretaceous and Tertiary 236 Summary and conclusions 237 Introduction. problem and conclusion. The present paper deals with the extensive series of fresh-water deposits of the northwest {i. e., broadly, the region east of the Rocky Mountains and between Wyoming and the valley of the Mac- kenzie River) comprising what is here considered as the Fort Union formation. It is shown that the Fort Union embraces more than has been commonly assigned to it. Conformably below the beds by some geologists considered as the true Fort Union occur dark-colored sandstones, clays and shales, which have often been incorrectly referred to the Laramie, or its equivalent, but which are stratigraph- ically and paleontologically distinct from the Laramie, and the con- tention is here made that these beds, which include the " Hell Creek HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS l8l beds"" and so-called "somber beds" of Montana, the "Ceratops beds" or "Lance Creek beds" of Wyoming, and their stratigraphic and paleontologic equivalents elsewhere, are to be regarded as con- stituting the lower member of the Fort Union formation, and are Eocene in age. Historical Summary. To all students of the late Cretaceous and early Tertiary forma- tions of the northwest, the Fort Union formation is a familiar term. The name was originally given by Dr. F. V. Hayden^ in 1861 to his great lignite group which: "Occupies the whole country around Fort Union, extending north into the British possessions, to unknown distances; also southward to Fort Clark." At the same time it was stated that the formation had also been observed under the White River group on the North Platte River above Fort Laramie, and on the west side of the Wind River Mountains. The beds were des- cribed as consisting of clays and sands, with round ferruginous con- cretions, and numerous beds, seams and local deposits of lignite. The fossil contents were very abundant, consisting of great numbers of dicotyledonous leaves, fresh-water shells of several genera, scales of Lepisosteus, together with bones of Trionyx, Emys, Compsemys, crocodiles, etc. The abundant fossil plants obtained by Doctor Hayden were submitted to Dr. J. S. Newberry for study, and his report on them was published in 1868.'' The plants were collected at various points on the Missouri River, at Fort Clarke, at Red Spring thirteen miles above, at Fort Berthold, at Crow Hills, one hundred miles below Fort Union, at the mouth of the Yellowstone, on O'Fallon's Creek one hundred miles above the mouth of the Yellowstone, and in the valley of that stream. On account of their association with invertebrates reported by Professor Meek to be of Miocene age, as well as from their agreement with plants stated by ^ In this paper the terms "Hell Creek beds," "Ceratops beds," "Lance Creek beds," "Somber beds," "Upper Laramie" and "Black Buttes beds" are placed within double quotation marks because these terms have not been formally adopted by the U. S. Geological Survey. In other cases, e. g., 'Fox Hills,' 'Laramie' etc., single quotation marks are used only when the beds are not regarded by the writer as properly identified. ^ Proc. Acad. Sci. Phila., 1861, p. 433. * Ann. N. Y. Lye, vol. 9, April, 1868, pp. 27-76. 1 82 KNOWLTON Heer to be of this age from the mouth of the Mackenzie River, in Greenland, Spitzbergen, and various European localities. Doctor Newberry was led to refer them to the Miocene. Subsequent events have shown, however, that the position assigned the above mentioned beds by Heer was too high, and as a consequence it has come to be recognized that the Fort Union beds are beyond question of Eocene age. This result, however, was not reached without much diver- sity of opinion and conflict of authority. It is largely to the efforts of Dr. J. S. Newberry that we are indebted for keeping alive the question of the distinctness of the Fort Union from the typical Laramie. From the beginning of his studies of the Fort Union flora, until his latest utterance on the subject, he insisted upon their separation. His last words concerning it are as follows:^ Whether the Laramie is Cretaceous and the Fort Union Tertiary are other questions, but they are certainly distinct from each other, distinct in the general botanical facies of their floras as well as in the absence of common species. That the Fort Union flora is Ter- tiary there can be no reasonable doubt; it has many species in com- mon with the recognized Tertiary in the Canadian provinces of North America, in Greenland, and in the British islands, and it contains some plants which are living at the present day. Moreover, the grouping of the plants comprising it gives it a facies which enables one to recognize it at a glance. The abundance of species of Populas, Viburnum, and Corylus, imparts to it an aspect as different from that of the flora of the Laramie as are the recent floras of Europe and America from each other. That it is of Tertiary age is no longer seriously questioned. Areal Distribution and Lithologic Character of Fort Union Formation. The Fort Union formation, as now known, covers a vast area in the central Canadian provinces and, as predicted by Doctor Hayden, touches the Arctic Ocean in the valley of the Mackenzie River, while to the southward it is the surface formation over much of the western ' Trans. N. Y. Acad. Sci., vol. 9, 1889, pp. 30, 31. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 183 half of North Dakota, eastern Montana as far west as the Bridger Range, western South Dakota, eastern and central Wyoming, and northwestern Colorado. It is a fresh-water formation, consisting of comparatively fine material, mainly clay shale, sands and soft sandstones, with numerous beds of lignite and occasional thin beds of impure limestone. Owing to the many alternations of soft rocks with thin hard layers extensive erosion has given rise to the well- known bad-land topography so characteristic of much of the area covered by the formation in Montana, North Dakota and Wyoming. The maximum thickness of the Fort Union exceeds 8000 feet. It rests, sometimes in apparent conformity and in other cases with unconformity, on various underlying formations. The relationship with lower beds will be discussed later. Division of Fort Union Formation into Two Members. In many cases — notably in the vicinity of Hell Creek, along the Yellowstone River at Miles City and Glendive, Montana, adjacent southwestern North Dakota, and east of the Bighorn Mountains in Wyoming — it is possible to separate the Fort Union into two mem- bers on the ground of a marked difference in lithologic character. The upper member is composed in the main of light-colored, yellow- ish sandstones and clays, while the lower member is made up of many alternating beds of clay shale and sandstone of a dark gray or somber hue, whence, as a convenient field term, they have often been designated the "somber beds." The contact of the lower and upper members is so sharp, and the contrast in coloration so marked, that they may usually be recognized at a distance and traced with little difficulty. In certain parts of the area a thick bed of coal, or a bed of red baked clay due to the burning out of the coal, marks the point of contact. The present paper deals only with the strati- graphic relations and paleontology of the lower member. The dark gray so-called "somber beds" of eastern Montana naturally attracted the attention of those who visited this region. From their stratigraphic position and the sharp lithologic contrast between them and the overlying beds, they were presumed to belong to the Laramie, but recently secured paleobotanical and strati- graphic data shows that they have little or no relation to the Laramie, x84 KNOWLTON and must be considered as the lower member of the Fort Union. So far as known to the writer the Laramie is not present in this general region. Areal Distribution and Paleontologic Contents of Lower Member of Fort Union Formation. i. hell creek, montana. In 1907 Mr. Barnum Brown" published a valuable contribution on the geology of the Hell Creek region of northeastern Montana. To the beds in this area which contain a rich dinosaur fauna he gave the name Hell Creek beds, and alFO stated that: These beds are exposed on the Yellowstone River at Sentinel [Cas- tle?] Butte near Forsyth; at Glendive; near Eklaka; and at Hockett P. O., south of the Yellowstone. They are probably continuous with the dinosaur-bearing beds of the Little Missouri, and of the Grand and Moreau Rivers. The "Hell Creek beds," which are between 300 and 400 feet in thickness, are stated by Brown to rest unconformably on the Fox Hills, and are composed of two members, a basal massive sandstone, and an upper member of alternating dark arenaceous clays, carbon- aceous shales and sandstones. Overlying the dinosaur clays, and forming an "uninterrupted continuation" of them, are the so-called lignite beds 100 feet in thickness which are characterized by the presence of numerous beds of coal; they are referred by Brown with question to the Fort Union. Above the lignite beds without appar- ent break are the light-yellow sandstones and clays of the "identi- fied Fort Union." At several points in the "Hell Creek beds," but particularly in the massive basal sandstone. Brown records the pres- ence of numerous concretions which are sometimes of very irregular shape, but often resemble huge tree-trunks. It is perhaps noteworthy that similar concretions occur in approximately the same stratigraphic position at widely scattered localities within this lower member of the Fort Union, as at Forsyth, Miles City, near Glendive, Montana, • Bull. Am. Mus. Nat. Hist., vol. 23, 1907, pp. 823-845. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 185 between the Grand and Moreau Rivers, South Dakota^ and Converse County, Wyoming. The fauna of the "Hell Creek beds" is a comparatively rich one, comprising a few mammals, numerous dinosaurs belonging to the families Ceratopsidos, Trachodontida?, etc., together with crocodi- leans, turtles, scales and vertebras of fishes, and some thirty species of invertebrates, mainly Unios. On the basis of the identity of many genera and species of both vertebrates and invertebrates, as well as on the striking lithologic similarity, the "Hell Creek beds" of Montana are correlated by Brown with the "Ceratops beds" of Converse County, Wyoming, a correlation confirmed by Hatcher and Lull in the recently published monograph of the Ceratopsia.' Up to the time of the publication of Brown's paper (1907), but few fossil plants had been found in the " Hell Creek beds, " and these in only the upper member. Dr. Arthur Hollick, to whom they were submitted, identified Equisetum lavigaium Lesq., Rhamnus salicifolius Lesq., Ficus spedahilis Lesq., Pterospermites sp., Sequoia heerii Lesq., and a mass of seeds resembling Sahalites friictifer Lesq. To these I was able to add a peculiar undescribed Ficus fruit, which has also been found in Converse County and at Forsyth. In the past season, however, Mr. Brown was fortunate in securing a fine collection of leaves in direct association with the skeleton of a dinosaur. This collection, which he has kindly permitted me to study, embraces the following species: Big Dry Creek, 60 miles south of Glasgow, Montana. Clays between middle and basal sandstones and in association with skeleton of a dinosaur. Sequoia nordenskinldi Heer. Taxodium occidenlale Newb. Ginkgo adiantoides (Ung.) Heer. Populus cimeata Newb. Populus amblyrhyncha Ward. Qucrcus sp. (Same as at Glendive.) Ficus artocarpoides Lesq. Sapifidus affinis Newb. East side of Big Muddy Creek, 28 miles south of Lisner, Montana. 250 feet above Fox Hills. Sequoia heerii Lesq. ' Todd, Am. Geol., vol. 17, 1896, pp. 347-349. * Mon. U. S. Geol. Surv., No. 49, 1907, p. 1S4. 1 86 KNOWLTON Ficus sp. (Same as found at Forsyth, Montana, and Ceratops beds of Converse County.) Viburnum antiquum (Newb.) HoUick. 2. MILES CITY, MONTANA AND VICINITY. In 1907 Messrs. A. J. Collier and C. D. Smith, while engaged in coal classification work, surveyed about 1000 square miles of country comprising the so-called Miles City, Montana, coal field. With the exception of some unimportant alluvial deposits in the valleys, all the rocks of the region were found to belong to the Fort Union formation, which they show is comprised of two members. The lower member, which corresponds to the "Hell Creek beds" just described, is about 500 feet in thickness, but the base is not exposed; however, from the comparison of the Miles City section with that at Glendive, about 70 miles to the northeast, it is inferred that the base is not very deeply buried. Collier and Smith^ give the fol- lowing detailed section of the lower member of the Fort Union for- mation at Miles City, which is here presented entire to show not only the varied character and coloration of the beds but the relative posi- tions of the paleontological material obtained. Section of the lower member 0/ the Fort Union formation near Miles City, Montana. ft. in. Limestone at base of up;:er member 3 Shale, white 7 Coal and shale, dark 6 Shale, gray 20 Limestone, weathering brown i Shale, light gray 10 Shale, dark gray 8 Limestone, weathering reddish brown 6 Shale, dark yellow 6 Coal 6 Shale, gray 21 Limestone, weathering light red i Shale, gray 25 Sandstone, white, with limes lone concretions 8 Shale 2 Sandstone 8 Shale, white 2 " Bull. U. S. Geol. Surv., 341A, 1908, pp. 39, 40. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 1 87 ft. in. Shale, gray 5 6 Sandstone, gray 20 Limestone, weathering red 3 Sandstone, gray 4 Limesione, weathering red 6 Sandsi one, gray 4 Coal and shale i 10 Sandsi one, gray 3 6 Limesione, weathering light red i 6 Shalt dark 3 Limestone, weathering red 3 Sandstone, gray 2 Coal 8 Shale, yellow 5 Limestone, weatherir-g light red 8 Shale, gray 13 Limestone, weathering red 6 Sandstone, fine gray 3 Limestone, weathering red 2 Sandstone, gray 2 Limestone, weathering red 2 Sandstone 5 Coal (Weaver bed) (D) 2 4 Shale 5 Coal I 3 Shale 10 Coal 6 Shale 6 Coal 3 Shale, sandy, dark yellow 6 Limestone, brittle 6 Shale, dark gray 5 Shale, fissile, dark brown 4 6 Coal 6 Shale, gray, with numerous limestone concretions 17 Limestone, weathering red 6 Shale, dark, with thin layers of i-ed limestone 10 Sandstone, brown, containing fragments of carbonaceous material 7 Coal I 6 Shale, light gray 9 Limestone, red 2 Shale 3 Coal 3 Shale, blue 8 Coal, with three small streaks of bone 3 4 Sandstone 6 Shale • 4 1 88 KNOWLTON ft. in. Sandstone 2 6 Shale 3 Coal 7 Shale 6 Coal, dirty, with several white partings, one-half inch to i inch thick (Laneybed) (C) 5 Shale I Sandstone 2 Shale 4 Limestone, weathering red 6 Sandstone, gray with some thin layers of shale 25 Limestone, weathering red 6 Shale, sandy, yellowish gray 8 Coal 8 Shale 3 Sandstone, friable, gray 20 Shale, sandy, many concretions 5 Limestone, weathering red 3 Sandstone, coarse, gray 17 Limestone, weathering light red 6 Sandstone 3 Shale, brown with blue bands 10 Sandstone 6 Coal and shale 2 Shale, sandy, gray, usually cross-bedded 50 Coal 6 Shale, sandy 5 Coal (kircher bed) (A) 5 465 3 The lower member is "conformably overlain by more homoge- neous rocks, mainly sandy shale, which as exposed by erosion are generally of a light-yellow color," thus making the typical upper member. Plants have been obtained from the lower member in this area as follows: Six miles east of Miles City and 15 feet above the Kircher coal (A of sec- tion, yielded: Glyptostrobw; europccus (Brongn.) Heer. Popidiis cuncata Newb. Populus amhlyrhyncha Ward. Viburnnvi sp. Bluffs of Yellowstone opposite Miles City, 1 15-125 feet above river (C of section') : M arcJmnlia pealei Knowlton. Onoclca scnsibilis ossilis Newb. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 1 89 Corylus amcricana Newb. Popuhis amhlyrliyncha Ward. Populns cuneata Newb. Populiis nebrascensis Newb. Poptdus nervosa dongala Newb. Populits rotundifolia Newb. Cormts ncwhcrryi Hollick. N dumbo sp. Trapa micro pliylla Lesq., as determined by Ward. Cocculus haydenianus Wa d. Nine miles west of Miles City, Mont. Near N. E. cor., T 7, R. 45. "C" coal. Onoclea sensibilis fossilis Newb. Sequoia nordenskiijldi Heer. Populns cuneata Newb. Populus daphnogenoides Ward. Populns speciosa Ward. Populns arctica Heer of Lesq. Belula new (nearest B. coryloides Ward). Viburnum ajidquuni (Newb.) Hoi. Miles City field, S. W. \ sec. 22, T. 8, R. 47, about 50 feet below the Big Dirty coal bed. Celastrus pterosptrnioides Ward. Cclastrus ahiifoUus Ward. Celastrus curvinervis Ward. ElcEodendron polymor phuni Ward. Sapindus randifnliolus Ward. Crewiopsis populi folia Ward. Grew.'opsis platanijolia Ward. Diospyros haguei Kn. (small leaf). Viburnum antiquum (Newb.) Hoi. Populus amblyrhyncha Ward. Poptdus sp. ? Klonders ranch, 18 miles east of Miles City; 100 feet below top of "somber beds." Onoclea sensibilis fossilis Newb. G yptostrcbus europccus (Brongn.) Heer. Taxodium occidentale Newb. Sequoia (cone). Poptdus cuneata Newb. Populus amllyrhyncha W^ard. Sapindus affinis Newb. Hicoria antiquorum (Newb.) Knowlton. Signal Butte, 5 miles east of Miles City, 225 feet above base of section: Taxodium occidentale Newb. Platanus raynoldsii Newb. Same locality as last but in upper member of "sombei beds:" Clyptostrobus etiropccus (Brongn.) Heer. Taxodium occidentale Newb. ipo KNOWLTON Sequoia nordenskioldi Heer. Corylus americana Walt. Corylus rostrata Ait. Betula sp. Populus accrifolia? Newb. Popidus genetrix? Newb. Hicoria antiquorum (Newb.) Knowlton. Hicoria sp. Sapindus grandifoliolus Ward. Planera sp. Celastrus ovatus Ward. At a point on Sheep Creek about 40 miles east of Miles City (Sec. 16, T. 6 N., R. 52 E.) Collier obtained from just above the "D" coal the following vertebrate remains which were determined by Mr. C. W. Gilmore of the United States National Museum: Champsosauriis sp. Crocodylus sp. Dinosaur fragments. Turtles. Ganoid fish. Also a mammal tooth which Mr. J. W. Gidley has identified as Chriacus? sp. On Powder River, about 12 miles above Hockett, Montana, and a few miles south of the above mentioned locality, Brown^" reports finding parts of a Triceratops skeleton in the dark shale near the level of the stream and states that other fragmentary dinosaur bones were observed. In the vicinity of the Chalk Buttes, 20 miles west of Eklaka, Montana, dinosaur and other vertebrate remains were ob- tained by an expedition from the Field Museum of Natural History of Chicago. I am informed by Mr. E. S. Riggs, who conducted the party, that the following forms were secured: A fine skull of Triceratops calicornis, two skulls of other species, and a number of other specimens, of Ceratopsia, also a carapace and plas- tron of Basilcmys sinuosa, and individual bones of Trachydon and Chajnpsosaurus. 3. FORSYTH, MONTANA. The characteristic dark sandstones and clays of the lower Fort Union are known to extend as the surface formation for a distance *" Bull. Am. Mas. Nat. Hist., vol. 22, 1907, p. 823. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 191 of nearly 20 miles along both sides of the Yellowstone River west of Miles City, and apparently continue uninterruptedly to Forsyth, a distance of about 40 miles. Just south of the town the rocks rise in a steep bluflf about 250 feet in height, and consisting mainly of sandy shales with three layers of indurated sandstones. The beds dip greatly to the northeast. In a sandy clay about 75 feet below the top the following rather fragmentary plants were collected: Fern, probably a Dryopteris. Quercus sp. Querciis viburnifolia? Lesq. Lauraceous leaf (same form found in white basal beds at Glendive, Montana). Aralia sp. Sapindus affinis Newb. In the valley of Porcupine Creek, about 6 miles west of Forsyth, the characteristic dark shales of the marine Cretaceous are exposed, and in the usual concretions and lenses of limestone a considerable number of typical Pierre invertebrates were obtained. About one- half mile to the eastward of the fossiliferous marine Cretaceous are the lowest members of the "somber beds," there forming low bluffs and resting on the dark clay shales. The lowest bed is a soft yel- lowish-white sandstone, above which it becomes massive and brown- ish in color and often weathers into towers and irregular pinnacles. The beds at this point add probably 100 feet or more to the base of the section as exposed just south of Forsyth, while Chimney Butte on the west side of the Yellowstone, about 6 miles east of Forsyth, exposes beds apparently higher than those above mentioned, making a total thickness of the "somber beds" in this vicinity between 400 and 500 feet. No plants were found in either the lower or higher beds north of the river, though at a point about 5 miles northwest of Forsyth, in the lower beds, vertebrate remains have been found, and it was prob- ably at this locality that the bones of a large dinosaur were collected some fifteen years ago by Prof. I. C. Russell. In association with the bones were numerous specimens of the undescribed Ficus fruit already mentioned as occurring in the "Hell Creek" and "Ceratops beds. " 192 KNOWLTON 4. CUSTER, MONTANA, AND VICINITY. Beyond the point indicated above the conditions to the west and southwest of Forsyth are not definitely known, though the marine Cretaceous extends westward nearly to Myers; but from observa- tions made from a train, it seems not improbable that the "somber beds" may be traced as far as Bighorn or Custer. In any event, beds known to be of this age were observed near the latter place by Mr. L. J. Pepperberg, of the United States Geological Survey, dur- ing the past season. The section, there several hundred feet in thickness and resting on one of the upper members of the Pierre, is made up of light colored sandstones and shales. At a point near the top of the exposed section the following small collection of plants was secured: Four and one-half miles southeast of Bighorn, Montana. Platanus nohilis Newb. Popiilus cuneata Newb. Sapindus sp. 5. BULL MOUNTAIN, MONTANA, AREA. The so-called Bull Mountain coal field" of Montana covers an area of about 750 square miles, lying mainly to the south of the Mus- selshell River and some 20 miles north of the Yellowstone River. On the southeast the lower member of the Fort Union presumably connects more or less closely with the area just mentioned near Custer, while to the northeast it will probably be found to extend directly into the Hell Creek area. The beds in question, which are composed of usually light colored sandstones, clays and carbonaceous shales, are about 2000 feet in thickness and rest upon Pierre shale. They have usually been referred on stratigraphic grounds to the Laramie, but there is little question of their belonging to the lower Fort Union, though plants have not thus far been reported from them. 6. MELVILLE, MONTANA, AREA. The area now to be considered lies cast of the Crazy Mountains and about 25 miles north of Big Timber. The geological structure " See Woolsey, Bull. U. S. Geol. Surv., 341 A, 1908, p. 60. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS I93 in this general region is somewhat obscure — to judge by the different results obtained by the several geologists who have visited it — and owing to the lack of sufficient paleontological data at critical points in the section, it appears that the problem is still short of a complete and wholly satisfactory solution. Thus in 1896 Mr. W. H. Weed" published the record of a section taken along Lebo Creek in which he recorded the presence of over 7000 feet of Livingston beds, above which was a thickness of 4000 feet of Fort Union. A few years later Mr. Earl Douglass" visited this region, and although within a few miles of the Lebo Creek section of Weed, he failed to note the presence of Livingston rocks, referring this part of the section appar- ently to the 'Fox Hills' and 'Laramie.' The same area was sur- veyed by Mr. R. W. Stone"^ and party in 1907, and while he recog- nized tentatively the presence of the Livingston it was presumed to be only a few hundred feet in thickness and is located in another part of the section. The succession as worked out by Stone is as follows: Above the highest marine Cretaceous (Bearpaw of the Montana) in the region is a series of red and greenish sandstones 200 to 460 feet in thickness which form a conspicuous ridge, and which was provisionally regarded as a part of the 'Laramie.' The suc- ceeding beds, 1000 to 2400 feet thick, of soft shales and sandstones of a light-gray color, are regarded as the upper part of the ' Laramie. ' At this point Stone says: The gray beds of the Laramie formation are overlain, possibly with unconformity, by somber-colored sandstone and shale which may represent the Livingston formation. Sufficient paleontologic evi- dence has not been obtained, however, to determine the limits of these stratigraphic units. The thickness of the 'Laramie' is approximately 5500 feet, above which comes nearly as great a thickness of Fort Union. This section as interpreted by the writer is as follows: Above the Bearpaw is a series of shaley sandstones, at least several hun- dred feet in thickness, that unmistakably belongs to the Livingston, as plants identical with those found near the base of this formation *^ Am. Geol., vol 18, 1896, pp. 201-21 1. '^ Proc. Am. Phil. Soc, vol. 41, 1902, p. 217. '^*Bull. U. S. Geol. Surv., 341A, 1908, pp. 76-89. 194 KNOWLTON southeast of Bozeman, Montana, have now been located," and more- over the matrix is characteristically that of the Livingston. Between this point and the base of the upper member of the Fort Union according to Stanton is a thickness of 3000 to 5000 feet of beds, a portion of the lower part of which belongs with little doubt to the Livingston, but the top of the Livingston has not been definitely placed, though its maximum thickness apparently exceeds 2000 feet. The dinosaurs occur in this thick series of beds above the Livingston, and the beds are probably referable to the lower member of the Fort Union, though no plants have been found until within approximately 1000 feet of the top, where the following species have been obtained: Wido'ecomb Bros, ranch, northeast of Melville, Montana [sec. 29, T. 6 N.. R. 16 E.] Salix sp. Populus amhlyrhyncha Ward. Popidus cuneata Newb. Populus genetrix Newb. Populus daphnogenoides Ward. Populus sp. Aralia notata Newb. Legumino sites arachioi 'es Les(|. Vilis xantlwlithensis Ward. Credneria daturccfolia Ward. Phyllites cupanoides Newb. Carpites sp. These plants occur about 1200 feet below the small mammals now being studied by Mr. Gidley and regarded by him as of Tor- rejon afifinities, though scattering remains of mammals have been found down nearly to the horizon of the plants. The nearest point at which dinosaurs occur is stratigraphically about 600 feet below the mammal horizon, according to the collector, while according to Stanton the main Triccratops horizon may be 2000 to 3000 feet below this point. '* When this collection was first studied it proved very puzzling, and although no species were positively identified, it was tentatively regarded possibly Laramie or older. Subsequently, however, large collections were made and studied from the Livingston with the result of showing that it clearly belongs to the latter. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS I95 7. RED LODGE, MONTANA. The Red Lodge coal field lies along the foot-hills of the Beartooth Mountains, on the extreme south-central border of Montana. The Fort Union beds in this area show the immense thickness of 8500 feet, the productive coals being confined to a middle portion only about 825 feet in thickness, below which are barren beds 5700 feet thick. According to Mr. E. G. Woodruff^' who reported on the coals of this region in 1908, the lowest member is composed mostly of yellowish sandstone and shale. Beginning 1650 feet above the bottom of the formation is a group of beds 1000 feet thick composed of vari-colored sandy shale, with a few beds of soft yellowish sandstone, numerous beds of carbon- aceous shale, and in the upper part a few coal beds. Above this group carbonaceous shale and coal beds occur at diminishing inter- vals as the productive measure is approached. The lowest point at which Fort Union plants have been found in the section above described is about 700 feet above its base, and on this account this basal 700 feet has sometimes been referred tenta- tively to the 'Laramie.' In the opinion of the writer it is an insep- arable portion of the lower Fort Union. The exact area covered by the lower Fort Union in this vicinity has not been definitely determined, though it is known to extend to the west of Red Lodge for at least 25 miles, where it was found by Mr. W. R. Calvert resting conformably on the Livingston forma- tion, while to the east it extends nearly to Bridger. To the south- east this area undoubtedly connects with the beds of this age exposed as a ring about the margin of the Bighorn Basin, Wyoming. Plants have been obtained in the Red Lodge area at the following locali- ties: About 8 miles west of Bridger. Montana. Taxodium occidenlale Newb. Platamis nohilis Newb. riatanus haydcnii Newb. Populus avihlyrhyncha Ward. Sapindiis grandifolioliis Ward. Laurus sp. ? Ficus sp. " Bull. U. S. Geol. Surv., 341A, igo8, p. 92. Proc. Wash. Acad. Sci., August, 1909. 196 KNOWLTON About 8 or Q miles west of Bridger, Montana. Plalanus nobilis Newb. Populus incequalii? Ward. About 25 feet below the last collection. Taxodium occidniale Newb. Populus cuneata ? Newb. Platanus sp. About 25 miles northwest of Red Lodge, near center sec. T)T), T. 6 N., R. 18 E. 300+ above base. Platanus raynoldsii Newb. Same area as last, sec. 36, T. 5 N., R. 18 E. Within 300 feet of contact with Eivingston beds. Aralia notata Newb. Populus sp. 8. GLENDIVE, MONTANA, AND VICINITY. Returning to the Miles City field, as already set forth, the con- ditions are known for 40 miles to the east. Down the valley of the Yellowstone identical conditions prevail to near the vicinity of Iron BlufT where, at the mouth of Sand Creek, some 12 miles south- west of Glendive, an anticlinal fold brings the marine Cretaceous to the surface. The exposure is about 100 feet in thickness above the level of the river, on both sides of which it may be observed. It is upper Pierre (Bearpaw) in age and contains a rich fauna which has been listed by Dr. T. W. Stanton.*^ This exposure, which is known to continue for some distance to the southeast, is thought by Prof. A. G. Leonard" to be connected with a similar area along the extreme eastern edge of Custer County, Montana, and adjacent counties in North Dakota. Immediately above the Pierre, and in apparent conformity, are 150 feet of beds the lower half of which is made up of sandstones and shales, and the upper half of a brownish sandstone. While these beds are in the position of the Fox Hills, their age is in doubt. No invertebrates have been found, the only fossils being two or three small collections of dicotyledonous plants which are so fragmentary as to be undeterminable. According to Leonard'" they do not " Bull. U. S. Geol. Surv., 316, 1907, p. 196. " L. c, p. 197. " Loc. cit., p. 197. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 1 97 agree with the Fox Hills beds of the Hell Creek section, and should probably be referred to the lower member of the Fort Union as exposed in this region. The absence of invertebrates and presence of plants in a measure supports this view. On this point Leonard says: The line of contact between the Pierre and the overlying dinosaur- bearing beds, while not discordant so far as structure is concerned, may possibly represent a time break in which most of the upper fresh- and brackish-water beds of the Cretaceous are wanting. Overlying the last mentioned beds is a very conspicuous white, massive sandstone 35 to 50 feet in thickness, which first appears at the north end of Eagle Bluff, 2 miles west of Glendive, and persists for many miles up Sand Creek. The upper surface is somewhat uneven, suggesting the possibility of an unconformity, though this may be only local and of little importance. In the middle of this white sandstone, at a point about 5 miles up Cedar Creek, Dr. A. C. Peale collected fragmentary plants, among them Populus cuneata Newb., while in the upper portion at Eagle Bluff the following forms were obtained: Ginkgo adiantoides (linger) Heer. Quercus sp. Lauraceous leaf. Ficus trinervis Kn. Fictis or Sapindus sp. Viburnum whymperi Hear. Viburmim n. sp. Other localities for plants in the Glendive area are as follows: Bluff east of Sand Creek, 6 miles above Glendive, Montana, 420 feet above the Pierre. Taxodium sp. Ginkgo adiantoides (Unger) Heer. Salix angusla Al. Br. Quercus breweri Lesq. Populus cuneata Newh. Mouth of Cedar Creek, iif miles above Glendive, Montana, 400 feet above Pierre. Glyptostrobus europceus (Brongn.) Heer. Sequoia sp. 198 KNOWLTON Popuhis genetrix NcM'b. Corylus rostrata Ait. Pterospermites minor Ward. Pterospermitcs wliitei Ward. Cocnilns haydenianns Ward. F.liTodcndron polymorphum Ward. Eight or nine miles above Glendive, Montana, 100 feet above Pierre. Taxodinm sp. Platanus raynoldsii Newb. Plataniis sp. Populus amhlyrhyncJia Newb. Corylus aniericana Walt. Carpolithes lineatus Newb. Bluffs of Yellowstone River i^ miles from Glendive," Montana, 400 feet above Pierre. Onoclca sensihilis fossilis Newb. Sequoia nordenskioldi Heer. Ginkgo adiantoides (U)iger) Heer. Bluffs \ mile east of town of Glendive, Montana. Base of bluff or 350 feet from top. Sequoia sp. Aralia notala Lesq. Same locality as last, 195 feet below top. Car piles sp. Populus sp. Same locality, 125 feet below top of bluff. Equisetum sp. Thuya interrtipta Newb. Sequoia nordenskioldi Heer. Ficus sp. Car piles sp. Bluff on west side of Yellowstone River opposite Glendive, Montana, about the same horizon as the last. Tlinya inlcrrupta Newb. Sapindus grandifoHolus W^ard. Bluffs along Yellowstone River at Harpster's ranch, 10 miles northeast of Glendive, Montana. In upper ] ortion of lower Fort Union. Equisclum sp. ? Clyploslrobus curopans (Brongn.) Heer. Populus cuneala Newb. Populus sp. Populus sp. Pla 'anus raynoldsii Newb. Plalanus guilhlvue Gopj). Sapindus grandifoliolus Ward. P( lygon fn n. sp. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 1 99 A few fragmentary remains of vertebrates have been found in the Glendive area. Thus, Barnum Brown" records having found the weathered fore Hmb of a Triceratops "in the badlands near Glen- dive at an elevation of about 50 feet above the railroad track," which would bring its position at about 275 to 300 feet above the top of the Pierre. He also adds that "several other fragments of Triceratops andTrachodont dinosaurs were seen in this locality but not sufficiently preserved for specific determination." In the bluffs just east of Glendive Dr. A. C. Peale and the writer secured, in 1907, fragmentary remains of turtles and a single mam- mal jaw near the base of the bluff, and remains of turtles and frag- ments of a large dinosaur at a point about 100 feet above the base. It is probable that more careful search in this vicinity would disclose the presence of other localities for vertebrate remains, of which we heard vague rumors from the residents. 9. GLENDIVE, MONTANA, TO MEDORA, NORTH DAKOTA. The lower member only of the Fort Union is exposed in the imme- diate vicinity of Glendive, the beds dipping at a very slight angle to the northeast, and it is 12 or 15 miles down the Yellowstone and some miles back from the river, before the yellow beds of the upper mem- ber appear. Thus in the valley of lower Seven Mile Creek, which enters the Yellowstone from the north about 10 miles below Glendive, one first passes from near the level of the stream over the dark sandy shales and clays of the lower member for a distance of 6 or 8 miles before the yellow sands and sandy clays of the upper member appear in the bluffs on either side. • Probably about 15 miles to the east (the exact point was not noted) of Glendive the lower member has disappeared and the upper member is the surface formation, a con- dition continuing probably to the valley of the Little Missouri River. Sentinel Butte, North Dakota, which rises 650 feet above the level of the plain in which it stands is entirely in the upper beds (except for a very thin capping of supposed Oligocene containing fish re- mains), as are at least the upper portions of the bluffs bordering the '* Bull. Am. Mus. Nat. Hist., vol. 23, 1907, p. 823. 200 KNOWLTON Little Missouri River. The valley of this stream, according to Leonard and Smith,2° has a depth of 420 to 440 feet below the sur- face of the plain in which it has been eroded, and may have cut into the beds of the lower member, though the deep well sunk at Medora carries the section down for a distance of 940 feet below the lowest beds exposed in the region, and this, on the basis of the thickness at Glendive, must be near to the base of the section. The entire thick- ness of the section in this area, including the beds passed through by the Medora well, is 1720 feet, and it is possible that the upper por- tion of the lower member of the Fort Union may be exposed, but as this is not definitely known it is omitted from present consideration. In any event all of the numerous finely preserved plants obtained in this region are distinctly of Fort Union age. 10. BISMARCK, NORTH DAKOTA, AND VICINITY. To the eastward of Medora along the line of the Northern Pacific Railroad Fort Union plants have been collected at a number of points, as at Lehigh, Dickinson, Sims, etc., but as the stratigraphic relations are not definitely known through this region, w-e may pass on to Bismarck, North Dakota, where the lower member is undoubtedly present. Near the mouth of Apple Creek, about 10 miles southeast of Bismarck, are exposures of soft sandstones and clays of dark-gray color which form a bluff 75 to 100 feet high. Near the base of this bluff at a point i^ miles above the mouth of Apple Creek the following very fragmentary plants were obtained: Adiantum? sp, Salix sp. Quercus sp. Ficus sp. Launis? sp. Carpites sp. The beds mentioned above are not the base of the lower member of the Fort Union, but at a pointonthcweslsideof the Missouri River about 20 miles below Mandan, Dr. T. W. Stanton obtained the fol- " Bull. U. S. Geo!. Surv., 341A, 1908, p. 15. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 20I lowing plants in shaley sandstone about loo feet above the top of the Fox Hills: Taxodium sp. Populus amhlyrJiyncIia Ward. Sapindus affinis Newb. Qiicrciis sp. Sassafras sp. (Same found in Ceratops beds of Converse County, Wyoming). Ficus sp. new ? Ficus sp. new. About 25 miles to the northward of Bismarck, near the town of Washburn, the yellow sands and sandstones of the upper Fort Union appear and become the surface formation as far at least as Coal Harbor. At old Coal Harbor in the bluffs overlooking the Missouri typical Fort Union plants and invertebrates were collected. II. YULE, BILLINGS COUNTY, NORTH DAKOTA, AND VICINITY. I am informed by Prof. A. G. Leonard that near Yule, in western Bowman County, North Dakota, the lower member of the Fort Union (dinosaur-bearing beds) again makes its appearance, and is there about 600 feet in thickness. He has kindly prepared the fol- lowing section: Section of lower member of Fort Union {dinosaur beds) in the vicinity of Yule, North Dakota. Typical buff and light-gray shales and sandstones of the Fort Union out- cropping on the Little Missouri in the vicinity of Yule, where they are seen resting on the "somber beds." Somber beds 75 to 100 Shales and sandstones carrying leaves 25 to 50 Somber shales and sandstones 200 to 250 Somber beds containing dinosaur bones. These lie as near as could be estimated, about 200 feet above ihe unconformity. The vertical range of the dinosaurs wa.s not determined. Somber shale and sandstones to unconformity 200 Unconformity Sandstone and andy shale; no fossils found in these beds, but they are perhaps Fox Hills 80 Pierre shale yielding abundant typical Pierre shells, exposed 60 202 KNOWLTON This section, it will be observed, is strikingly similar to that at Glendive, Montana, and it is extremely probable, as already pointed out, that it is the same anticlinal fold which brings up the marine Cretaceous in both areas. From the shales and sandstones indicated as carrying leaves in the above section, the following species were obtained: Taxodium occidentale Newb. Popidns amllyrhyncha Ward. Pla anus haydenii Newb. Juglans riigosa? Lesq. Hicoria anliqitorum (Newb.) Kn. Sapindus affinis Newb. Viburmtm wliyniperi Heer. Trapa micropliylla Lesq. of Ward. Cocculus Juiydenianus Ward. Professor Leonard has sent specimens of the dinosaur obtained from near Yule to the United States National Museum, and it has been identified by Mr. Gilmore as Triceratops horridus (Marsh), this species being the type of the genus. 12. PROBABLE AREA OF LOWER FORT UNION IN NORTHWESTERN SOUTH DAKOTA. The exact outline of the area covered by the lower member of the Fort Union in southwestern North Dakota is not at present definitely known, nor has its extension to the southward been thoroughly traced, though I am informed by Mr. J. W. Gidley that he has ob- served dinosaur-bearing beds that are stratigraphically and litho- logically similar to the "Ceratops beds" of Converse County, W^yo- ming, in the country between the Grand and Moreau rivers in South Dakota, on both the east and west sides of Long Pine Hills. From these beds near the eastern base of Long Pine Hills remains of gigan- tic Trachodont dinosaurs have been exhumed. Since the above was written Mr. W. R. Cahert has sent in (under date of May 2, 1909) the following collection from McCor^l coal bank in N. W.i, Sec. 5, T. 129 N., R. 88 W., on the north side of the Cannonball River in the Standin" Rock Indian Rescr\ation. The HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 203 horizon is about 250 feet above the Fox Hills and embraces the following list of species: Thuya inter rupta Ncwh. Sequoia nordenskioldi Hecr. Populus avtblyrliyncJia Ward. Populus daphnogenoidcs Ward. Sapindus grandifoliolus Ward. Aristolocliia cordi/olia Ncwb. Cclastrus curoinervis Ward. Pali rus pealei? Ward. Viburnum sp., cf. V. antiquum (Newb.) Hoi. 13. AREA IN WESTON COUNTY, WYOMING. It is quite probable, however, that the beds under consideration can be traced from the area last considered in more or less contin- uous exposures into Weston and Converse counties, Wyoming. In the southwestern portion of the Newcastle quadrangle which lies mainly in the extreme southwestern corner of Weston County, Mr. N. H. Darton^* has described as ' Laramie' the beds in question which overlie the Fox Hills. Although Darton does not mention an unconformity between the Fox Hills and the 'Laramie, ' he speaks of the difficulty of drawing the line between them. However, Mr. Barnum Brown found uncon- formable relations between them. He says:^^ On Alkali Creek, about 35 miles northwest of Edgemont, S. D., and 6 miles north of the Cheyenne River the dinosaur-bearing beds do rest on the marine Fox Hills. In 1901 I obtained characteristic fossils from both formations near their contact at that locality. At that place the conditions are similar to those in the Hell Creek region. From the base of the "series of lignite beds overlying the Converse County beds similar to those overlying the Hell Creek beds" on Seven Mile Creek, 40 miles northwest of Edgemont, Brown obtained the following plants: Taxodium occidenlale Newb. Sequoia nordenskioldi Heer. ^' U. S. Geol. Surv., Folio 120, 1904. " Bull. Am. Mas. Nat. Hist., vol. 23, 1907, p. 844. 204 KNOWLTON Dammara? sp. Plalanus raynoldsii Newb. Platanus sp. Quercus sp. Car piles sp. From the same locality on Seven Mile Creek, "but lower and as- sociated with a dinosaur skeleton (Claosaurus annectens Marsh)," the following plants were collected by Brown: Sequoia heerii Lcsq. Taxodium distichum miocenum (Brongn.) Heer. Ginkgo adianioides (Ung.) Heer. Musophyllnjn sp. (probably new). Flahfllaria eocenica Lesq. Sal allies grayanus Lesq. Palmocarpon palmarum (Lesq.) Kn. Plalanus rJiomboidca Lesq. Plalanus sp. Ficus speclabilis Lesq. Viburnum sp. 14. NORTHWARD EXTENSION OF WESTON COUNTY AREA CONNECTING WITH THE MILES CITY AREA. It is now possible, through the work of Mr. E. S. Riggs of the Field Museum of Natural History, to extend the known distribution of the dinosaur-bearing beds to the northward of the Weston County area, and make practical connection with the beds of similar age in the Miles City, Montana, field. Mr. Riggs informs me that on passing northward from New Castle, Wyoming, these beds were first encountered on the head waters of the Little Missouri River 20 miles west of Devil's Tower. Again, on the east fork of Little Powder River, in Montana, he found a weathered skeleton of Trachy- don, partial skulls of Ccralopsia and fragments of a large carnivorous dinosaur, probably a Tyrannosaurus. The formation was thence traced along the east bank of Powder River from Powdcrvillc, Mon- tana, to a point on Sheep Creek some miles northeast of Mizpah, which brings it well within the area studied by Collier and Smith. During the past season (1908) Mr. R. W. Stone of the U. S. Geolog- ical Survey, found remains of Ccralopsia in the vicinity of Moore- croft, Wyoming, which makes an additional j3oint connecting the areas. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 205 15. CONVERSE COUNTY, WYOMING, AREA. We may now take up the consideration of the celebrated "Cera- tops beds," a direct northern extension of which are those just mentioned in Weston County. In 1889 Marsh^^ gave the name of " Ccralops beds" to a series of beds characterized by the presence of an extensive fauna of the then little known group of quadrupedal, horned, herbivorous dino- saurs. He did not then, nor indeed subsequently, give any very definite account of either the geographic location or stratigraphic position of the beds, beyond stating that they were in the 'Laramie of Woyming,' and that the horizon "has now been traced for nearly 800 miles along the eastern flank of the Rocky Mountains. " Of the position he says: They are fresh-water or brackish deposits, which form a part of the so-called Laramie, but are below the uppermost beds referred to that group. In some places, at least, they rest upon marine beds which contain invertebrate fossils characteristic of the Fox Hills deposits. It was reserved for Professor Marsh's assistant, Mr. J. B. Hatcher, to give the first definite and precise information on these points. The principal, and what may be called the typical, "Ceratops beds" are really of very limited extent, occupying a strip about 15 miles in width from east to west, by 30 miles in length from north to south, in the northeastern part of Converse County, Wyoming, along Lance Creek and the area drained by the lower portions of Docgie, Cow, Buck, and Lightning creeks. This area furnished not only all of the Ceratopsidas described by Marsh from Wyoming, but fully 95 per cent of the entire group known at that time. The "Ceratops beds" are best exposed along the eastern and southern borders of a synclinal basin, and according to Hatcher are 3000 feet in thickness, though Dr. T. W. Stanton and myself, when we visited the area in 1896^^ concluded that they could hardly exceed 2000 feet, but as a large portion of the beds are exposed at a low angle in a broad, flat, grassy plain, it is impossible to measure the beds with a great degree " Am. Jour. Sci. (3), vol. 38, 1889, p. 501. "* Bull. Geol. Soc. Am., vol. 8, 1896, pp. 128-137. 2o6 KNOWLTON of accuracy. The entire section of the region, which begins with several hundred feet of soft, bluish shales of the Pierre, up to and including the acknowledged Fort Union, was supposed by Marsh and Hatcher to be one of continuous deposition; that is to say, no actual unconformity had been detected. The Fox Hills, with an estimated thickness of 500 feet, consists of an alternating series of sandstones and shales. The massive sandstones at the top contain numerous large concretions and a rich marine fauna of character- istic Fox Hills species. The line between the Fox Hills and the overlying beds is a difficult one to draw. Hatcher, at first, placing it arbitrarily at a six-inch band of hard sandstones which separates the fossil-bearing Fox Hills sandstone below from the very similar but non-fossiliferous sandstones above. Later, however, Hatcher appears to have changed his mind regard- ing the lower limits of the "Ceratops beds," for he says:-^ At no place in the Converse County region do the true Ceratops beds, with the remains of horned dinosaurs, rest upon true marine Fox Hills sediments; nor are the Ceratops beds in this region over- lain by strata which could be referred without doubt to the Laramie. This point was apparently well taken, for Stanton and 1 found four species of brackish-water invertebrates in clays above a forty-foot bed of massive sandstone over 400 feet above the highest fossiliferous Fox Hills horizon in that particular section. The fact remains, however, that the fossiliferous portion of the "Ceratops beds" is mainly the upper portion, the highest point at which dinosaurs were found, being only 100 to 150 feet below the Fort Union. The fossil-bearing members of the "Ceratops beds" consist, as described by Hatcher,^' of alternating sandstones, shales, and lignites, with occasional local deposits of limestones and marls. The diderent strata of the series are not always continuous, a stratum of sandstone gi\ing place to one of shales, and vice versa. This is especially true of the upper two-thirds of the beds. This lack of continuity in the different strata has rendered it well nigh impossible to establish any definite horizons in the upper members of the series. "Am. Nat., vol. 30, 1896, p. 117. *• Am. Jour. Sci. (3), vol. 45, 1893, P- ^37- HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 207 Following is the list of plants thus far obtained from the " Ceratops beds" of Converse County: Alga (Gen. et sp. nov.) Aralia sp. (Like species from Forsyth). Car piles 2 sp. Cyperacites sp. Equisetum sp. ? Ficus 5 sp. Flabellaria cocenica Lesq. Grewiopsis cocenica (Lesq.) Kn. Fern (Gen. et sp. nov.). Hicoria sp. Juglans 2 sp. Lysimachia sp. (new). Myrica torreyi Lesq. Oitelia? sp. (new). Pahnocarp.:n palmarum (Lesq.) Kn. PhyUites 3 sp. Platanus raynoldsii Newb. Platanus raynoldsii integrifolia Lesq. Populus subrotundala Lesq. Popidus amhlyrhyncha Ward. Quercus rinereoides Lesq. Quercus viburnij'olia Lesq. Quercus 3 sp. (new?). Saial rigida Hatcher. Salix angusta Al. Br. Salix 3 sp. (new?). Sa'vinia sp. Sassafras sp. (New; same found below Mandan, N. D.) Sequoia nordenskioldi Heer. Taxodium distichum miocenum Heer. Trapa '.nicrophylla Lesq. Ulmus sp. Viburnum sp. (new'^. Viburmini whymperi Heer. (Also found at Glendive, Hell Creek, and below Mcndan. In 1896 Dr. T. W. Stanton and the writer spent about three weeks in the study of the " Ceratops beds" of Converse County, our results being set forth in a joint paper under the title: "Stratigraphy and Paleontology of the Laramie and related formations in Wyoming."" After reviewing the stratigraphy and local development in this region, so far as our observations permitted, the "Ceratops beds" were " Bull. Geol. Soc. Am., vol. 8, 1897, pp. 127-156. 2o8 KNOWLTON correlated with the dinosaur-bearing beds at Black Buttes, Wyo- ming, and both were then concluded to be of Laramie age. For several years past, however, data having a distinct bearing on the problem have been accumulating in adjacent and more remote areas, which have indued the present writer to completely change his opinion concerning the relations and age of the beds in question. When the invertebrate fauna and the flora of the " Ceratops beds" are compared with those of Black Buttes alone, it is true that a certain degree of relationship can be established. Among the invertebrates, for instance, it was shown that 9 species — 7 of which belong to Unto — of the 28 species found in the "Ceratops beds," occur also at Black Buttes, while of the 24 forms of plants then recognized it was possible to identify only 8 specifically, one of which was from an unknown locality, thus leaving less than one-third of the total num- ber on which to ascertain the bearing on the question of age. Of the remaining 7 species, 3 were found at Black Buttes, one in the Montana at Point of Rocks, 2 in the Laramie at Golden, Colorado, and one in the Raton Mountains of New Mexico, then supposed to be of Laramie age but now thought to be higher. Of the undescribed forms I wrote at that time as follows: The afiinity of the undescribed forms is also quite clearly with the true Laramie flora, and thus as nearly as can be made out, the plants confirm the Laramie age of the Ceratops beds. This statement was based on tenative comparisons with beds then supposed to be Laramie in age, but which subsequent investigation has shown belong to higher horizons. Since the publication of the above-mentioned paper the plants have been more thoroughly stud- ied, with the result of increasing the number to 48 forms, as already listed, though still showing a large proportion of new forms, with more modern aflmities. Of the 14 species having an outside distribu- tion, 9 are found in the Fort Union, 5 in the Shoshone, and 4 each in the Laramie and Montana, thus slightly modifying the distribu- tion as given in Brown's paper.^^ On these grounds the correlation of the "Ceratops beds" with the beds at Black Buttes is no longer admitted by the writer, and the Laramie age of the "Black Buttes beds" is held in abeyance. " Bull. Am. Mus. Nat. Hist., vol. 23, 1907, p. 844. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 209 The vertebrates and invertebrates found in the typical " Ceratops beds" will be considered later. 16. AREA TO THE EASTWARD OF THE BIGHORN MOUNTAINS, WYOMING. To the northwest of Converse and Weston counties, Wyoming, the beds under consideration occupy a vast area in the central Great Plains region between the Black Hills and the Bighorn Mountains, extending into Montana for an unknown distance, and completely encircling the Bighorn Basin. The complete areal distribution in this region is not yet a matter of published record, though the pres- ence of the beds at numerous points is attested. Mr. N. H. Darton^^ to whom we are indebted for most of the published knowledge con- cerning this area, made the following disposition of the Upper Cre- taceous and Tertiary rocks of the region. The name of Parkman sandstone was given to the several hundred (about 350) feet of soft sandstone overlying the more typical Pierre shale, which, from its position, was supposed to represent the Fox Hills, but according to Dr. T. W. Stanton it may be as old as the Claggett — one of the lower members of the Pierre. Immediately above the Parkman but separated from it with difficulty, is the Piney formation, a name proposed for the lowest formation of the thick series of fresh-water sandstones and shales of later Cretaceous age, formerly designated 'Laramie,' lying in the great Basin adjoining the Bighorn uplift. The maximum thickness of the Piney, according to Darton, is about 3000 feet, but according to Mr. C. A. Fisher this does not constitute a lithologic unit, only the lower 600 to 800 feet of sandstones being properly referable to this formation, while the remainder, composed of gray sandstones, dark shales and coal streaks, belongs to the lower Fort Union (somber beds). Whatever its proper limits, so far as present facts indicate, the whole section above the marine Creta- ceous is to be regarded as belonging to the Fort Union. Unconformably overlying the Piney as defined by Darton, but probably without the intervention of any considerable erosional interval, is a great thickness of conglomerate to which the name " Geol. Bighorn Mts., U. S. Geol. Prof. Paper, No. 51, 1906. 2IO KNOWLTON Kingsbury conglomerate has been given. It occupies a narrow band only 6 or 8 miles in width and 25 to 30 miles in length, adjacent to the mountains, and strikes approximately north and south. Its north and south extent coincides very nearly with faults which, accord- ing to Darton, attended an uplift of some 9000 feet of this portion of the Bighorn Mountains. This uplift, which involved also the Piney formation, resulted in a sharp fold, or in some cases probably a break, of the underlying beds resting against the granite mass, while the increased elevation of the mountain mass accelerated ero- sion and precipitated the debris which forms this conglomerate, embracing as it does rocks of all ages from the granites to and includ- eng the Piney. That the unconformity at the base of the Kings- bury is not of wide significance is shown by the fact that it was not detected in a section made at Parkman, only about 25 miles north of the northernmost extension of the Kingsbury, nor has it been found a few miles to the southward of the southern limits of the formation. Along the eastern margin of the Kingsbury it may be observed finger- ing into the soft shales and sandstones of the De Smet formation, the name given by Darton to the 5000 feet or more of beds overlying the Piney, and with which sedimentation was apparently continuous; in fact the Kingsbury was regarded by Mr. J. A. Tail, who surveyed this area for coal in 1907, as merely the near-shore phase of the upper member of the Fort Union. At a number of horizons in the Kings- bury Fort Union plants have been collected, and at a point about 650 feet above its base Mr. H. S. Gale obtained a mammal jaw which Air. Gidlcy identifies as identical with a Fort Union species found at Fish Creek, Montana. The first two of the following lists of plants are from the Kings- bury conglomerate, and are given to show how similar is the flora of the upper and lower members of the Fort Union; the others are from the lower member: Sequoia sp. ? Cyperarites sp. Populus sp. ? Platanus haydenii Newb. Ficus plani ostata? Lesq. Cinnavwmum a (fine Les(|. Sapindiis grandifolioliis Ward. Diospyros ficoidea Lesq. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 211 Cissiles parroticcfolia Lesq. Bcrchemia miiUinervis Al. Br. Aristolochia cordifolia Newb. Cornus ncwhenyi Hollick. Viburnum sp. Viburnum whympcri Heer. Carpites sp. Seven miles southwest of Buffalo, Wyoming. Dryophyllum suhfalcatiim Lesq. Ficus planicostata Lesq. Diospyrosjicoidea Lesq. Carpites sp. On road from Buffalo to Klondike, Wyoming. [Sec. 6, T. 49 N., R. 82 W.| Piney formation. Sequoia nordcnshioldi Heer. Po pubis arctica Heer of Lesq. Popidus amblyrhynclia Ward. Populus hobilis Newb. Sapindus grandifoliolus Ward. About i^ mile northeast of Dayton, Wyoming. Glyptostrobiis europcsus. Populus arctica Heer. Platanus guillelmcB Gopp. Platanus sp. Fruit, gen. et sp. ? Two and one-quarter miles northeast of Ranchester, Wyoming. Lower De S met or Piney. Taxodium occidentale Newb. Sequoia sp. ? Sparganium stygium Heer. Popiuus acerifolia Newb. Platanus nobilis Newb. Sapindus affinis Newb. Acacia sp. [new]. Bauhinia sp. [new]. Nelumbo sp. ? Two and one-half miles southeast of Ranchester, Wyoming; 75 feet below top of lower member. • Platanus sp. Hicoria antiquoruni Newb.) Kn. Sapindus grandifoliolus? Ward. Slater Creek, 6 miles northeast of Monarch, Wyoming; 40 feet below top of lower member. Populus aniblyrhyncha Ward. Popidus genetrix Newb. Populus sp. ? Proc. Wash. Acad. Sci., August, 1909. 212 KNOWLTON Platanus nohilis Newb. Legumino sites arachioides Lesq. Sapindus sp. Viburnum whymperi Heer. 17. BIGHORN BASIN, WYOMING. The following discussion of the lower Fort Union in this area is based on the published work of Mr. C. A. Fisher's,^" personal obser- vations made during the seasons of 1907 and 1908, and the work of United States Geological Survey field parties under the charge of Mr. E. S. Woodruff, made during the same years. In the extreme northwestern part of the Basin the lower Fort Union undoubtedly connects with the area about Red Lodge, Montana. At the time Mr. Fisher's paper was prepared sufficient paleontologic data had not been obtained to fix definitely the limits of this, as well as under- lying and overlying beds, and the whole series, aggregating more than 5000 feet in thickness, was described and mapped as ' Laramie and associated formations.' As shown on Fisher's map, these beds are exposed continuously about the outer edge of the Basin, having their maximum areal extent in the northeastern portion and their minimum exposure along the northwestern border just below the Montana state line. The lowest member of the aggregated ' Laramie and related forma- tions,' resting upon the Pierre, is composed of about 300 feet of a "massive gray to buff sandstone," which lithologically resembles the supposed 'Fox Hills' on the eastern side of the Bighorn Mountains, and for this reason was tentatively referred to the 'Fox Hills. ' The few invertebrates do not serve to fix definitely the age, according to Stanton,'* since they may occur below the Judith River as well as in Fox Hills. Above these beds, whether with or without an erosional interval is not definitely known, comes the great thickness of Fort Union which Fisher has shown may be divided into two parts. The lower member, which corresponds to the beds already so extensively traced, is composed of dark sandstones, shales and clays with occa- sional seams of coal and coaly shales. Throughout the lower por- *" Geol. and Min. Res. Bighorn Basin, Wyoming: U. S. Geol. Surv., Prof. Paper 53, 1906, p. 31 et seq. *' In Fisher, Bull. U. S. Geol. Surv., Prof. Paper, 53, p. 32. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 213 tion the sandstones are usually fine-grained and massive, and sand- stone concretions are of frequent occurrence. The line between the lower and upper members is a lithologic one, the sedimentation being continuous. At various localities in the Basin vertebrate remains have been obtained in this lower member, but they are usually so fragmentary that little can be said concerning them beyond the fact that they be- long to dinosaurs. Plants are abundant in many places, as the fol- lowing lists attest. The following localities for plants are known within the Bighorn Basin: Near I o P. O., Bighorn County, Wyoming; above 500 feet above base of beds. Onodea sensibilis fossilis Newb. Sequoia nordenskioldi Heer. Sequoia sp. Protophyllocladus? sp. new. Popidus cuneata Newb. Populus daphnogenoides Ward. Popidiis arctica Heer of Lesq. Populus n. sp. Populus amblyrhyncha Ward. Juglans sp. Corylus americana Wa'ter. Platanus haydenii Newb. Sapindus grandifoHoliis Ward. Celaslrus taurinensis Ward. Celastrus sp. Paliurus pulcherrimus Ward. Pal urus sp. Diospyros sp. Two miles north of Ilo P. O., Bighorn County. Ginkgo adiantoides (Ung!) Heer. Populus amblyrhyncha Ward. Quercus sp. Platanus nobilis Newb. Magnolia tenuinervis? Lesq. Ficus sp. (type of F. planicostaia). Celas rus sp. Berchemia midtinervis Al. Br. Viburnum newberryanum Ward. Viburmim antiquum (Newb.) Hollick. Twelve miles west of Ilo P. O., Bighorn County. Cocculus havdcnianus Ward. 214 KNOWLTON West of Winchester Hills, Bighorn Basin (Sec. 34, T. 45 N., R. 95 W.). Platanus nohilis Newb. Platamis haydenii Newb. Sapindus affinis Newb. Sapindus grandifoliolus Ward. Three miles northwest of Meeteetse, Wyoming. Black Diamond Mine. About 50 feet above base of section. Dryopteris sp. new. Sequoia langsdorfii or near it. Sapindus grandifoliolus Ward. Lauraceous leaf, probably new. Shoshone River, near Cody, Wyoming. [This and the following collection may be from the upper member.] Sequoia langsdorfii (Brongn.) Heer. Populus glandulifera? Heer. Platanus nobilis Newb. Sapindus grandifoliolus Ward. Jiiglans sp. Near Cody, Wyoming [Sec. 3, T. 53 N, R. loi W.] Platanus haydenii? Newb. Populus speciosa Ward. Hisoria antiquorum (Newb.) Kn. Shoshone River at mouth of Sage Creek, 3 miles below Cody, Wyoming. Platanus raynoldsH Newb. Platanus haydenii Newb. Platanus nobilis Newb. Southeast of Rairden, Bighorn Basin [Sec. 5, T. 48 N., R. 91 W.] Glyptostrohus europaus (Brongn.) Heer. Sapindus grandifoliolus? Ward, Sapindus affinis Newb. Populus sp. Bud Kimball mine, Bighorn County, Wyoming [T. 45, R. 89 W.] Glyptostrohus europaus Heer. Taxodium occidentale Newb. Viburnum whymperi Heer. Three-fourths mile west of Nowater mine [T. 44 N., R. 90 W.], Bighorn County, Wyoming. Populus cuneata Newb. Platanus raynoldsii Newb. Hicoria antiquorum (Newb. )Kn. Carpites sp. Bluff on east side of Bighorn River, opposite Kirby, Wyoming; 200 feet above base of lower Fort Union. Glyptostrohus europceus Heer. Sequoia nordenskioldi Heer. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 215 Taxodium occidentale Newb. Platanus guillelnKC Gopp. Cocculus liaydcnianus Ward. Same locality as last but about 300 feet above base of lower Fort Union. Taxodium occidentale Newb. Populus amblyrhyncJia Ward. Popnlus cuneata? Newb. Populus genetrix Newb. Viburnum perplexum Ward. Same locality as last two, but near top of lower Fort Union. Taxodium distichum miocenum Heer. Berchemia multinervis Al. Br. Celastrus ferrugineus Ward. 18. POSSIBLE DISTRIBUTION OF LOWER FORT UNION IN OTHER AREAS. Having passed in review the areas where the lower member of the Fort Union is known to occur, a brief mention may be made of certain extra-limital areas where the evidence is incomplete or con- flicting, and further data are to be looked for. In the central Canadian provinces the Fort Union is known to be present over a wide area, and it is more than probable that it will ultimately be found possible to differentiate the lower and upper members; in fact it is thought that the Edmonton beds of the Canadian geologists may correspond to the lower member and their Paskapoo beds to the upper member. But the writer has seen no material from Canadian sources, and the above reference is tenta- tive. From scanty information at hand it seems likely that the lower Fort Union will be found spreading over a considerable area in northern Montana near Big Sandy, along the Missouri River in the Fort Peck Indian Reservation and in southern North Dakota and northern and northwestern South Dakota. At Black Buttes, Wyoming, beds containing a Fort Union flora have been found resting with marked unconformity on the so-called ''Black Buttes beds" (the beds containing Agathaumas sylvestris), while 20 miles to the southward they rest on the Lewis shale, this condition continuing to the outliers of the Uintas near the Wyoming- Colorado line. That these beds with a Fort Union flora are the beds described and mapped by King as Vermilion Creek or Wasatch, 2l6 KNOWLTON is perhaps a sufficient excuse for omitting their further consideration until additional information, both paleobotanical and stratigraphic, is at hand. To the eastward of Black Buttes, in the vicinity of Creston, Riner and Rawlins, and as far at least as the vicinity of Carbon, Fort Union plants and occasional dinosaurs have been found. The beds containing them belong to the so-called "upper Laramie" of Veatch and others, that is above the unconformity at the top of the Laramie, but whether or not this series of beds is a unit is an open question at present. It seems probable to the writer that the lower portion may belong to the recently established Shoshone group of Cross, and the upper portion only to the Fort Union. To the west in the vicinity of Evanston a few Fort Union species have been noted, but in all these cases further data are demanded. Relations Between the Lower Member of the Fort Union AND Underlying Formations. Having traced the areal distribution of what is here called the lower member of the Fort Union formation, as completely as present facts seem to warrant a brief recapitulation of the relations that have been demonstrated of these to the underlying beds may be made. In the Hell Creek region of Montana, the vicinity of Yule, North Dakota, and on Alkali Creek, Weston County, Wyoming, the beds rest unconformahly on the Fox Hills. At Forsyth, near Custer, the areas south and east of the Bull Mountains, and probably at Glendive, all in Montana, and at Buffalo, Wyoming, the beds rest directly on the Pierre, and not always its uppermost member. In Converse County, Wyoming, the basal conditions are not definitely known, though presumably the relations may be similar to those obtaining in adjacent Weston County. Throughout the Great Plains area the Fox Hills is usually, but not always present, and even when present has not always been satisfactorily separated from the overlying beds. The evidence is conclusive where actual unconformity has been shown, as well as where the beds are found resting on a lower mem- ber (Pierre) of the Upper Cretaceous, and hence, in intermediate areas where discordance has not been observed, it is reasonably cer- HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 21'J tain that a time interval is represented during which certain of the Upper Cretaceous sediments and, if my view be correct, also certain lower Tertiary beds, including the Arapahoe and Denver or their equivalents, were removed. It is therefore demonstrated beyond reasonable doubt that the beds under consideration are the same throughout the wide area over which they have now been traced. As they are above an unconformity they can, on the basis of stratigraphy, no longer be considered as a part of the 'conformable Cretaceous series.' Relations Between Lower and Upper Members of the Fort Union. With the exception of the limited area covered by the Kingsbury conglomerate east of the Bighorn Mountains, which as already shown was a comparatively local affair consequent upon the Bighorn up- lift, the lower member of the Fort Union is conformably overlain by the upper member throughout the entire region. Of the many workers who have observed the field relations at hundreds of points, not one, so far as known to the writer, has recorded the presence of unconformity between them. The strong lithologic difference, as already described, makes it possible to separate them usually with little difficulty, but the sedimentation appears to have been contin- uous. The possibility of unconformity by overlap, always a diffi- cult condition to demonstrate, has not been overlooked, but if present it should have been detected on the borders of the areas along the mountains, which does not seem to have been done. Therefore, it appears that the lower and upper members cannot be separated on structural grounds. Stratigraphic Relations of the Lower Member of the Fort Union to the Laramie. The beds that are here regarded as constituting the lower member of the Fort Union have usually been — and by many are still — called 'Laramie,' but when it is recalled that, by original definition, the Laramie is the uppermost member of the conformable Cretaceous series above the Fox Hills, it is seen at once that they cannot be so considered since they are separated from the Cretaceous by an uncon- 2l8 KNOWLTON formity. Mr. Barnum Brown ^^ has clearly recognized the validity of this position, concerning which he writes as follows: Strictly following King's definition of Laramie, neither of these deposits [''Hell Creek beds," "Ceratops beds." etc.] can be con- sidered as such, for neither one represents a continuous sedimen- tation from the marine Fox Hills. They should therefore be grouped with the Livingston, Denver, and Arapahoe beds and may be considered Post-Laramie. This interpretation by Brown makes the unconformity at the base of the lower member of the Fort Union the same as that demonstrated by Mr. Whitman Cross as occurring at the top of the Laramie in the Denver Basin of Colorado. In the latter area, while an unknown thickness (estimated by Cross at 12,000 to 15,000 feet) has been removed, a considerable thickness of Laramie beds still remain, whereas in the areas covered by the lower Fort Union, the Laramie, as well as the Arapahoe and Denver, or their equivalents, if ever present, has been entirely removed so far as known, and the beds rest on other members of the Upper Cretaceous series. A comparison of the conditions of sedimentation in the two sets of beds, as indicated by their invertebrate fauna, confirms their distinctness. Thus, the Laramie is described as a series of brackish- and fresh-water beds, indicating transition from estuarine or marine conditions, while the Fort Union was laid down in fresh water, and affords little or no convincing evidence of even temporary or occa- sional incursions of the sea.^^ Paleontological Characters of the Lower Fort Union. I. plants. We may now proceed to the consideration of the various lines of paleontological data, beginning with that of the plants. On review- ing the foregoing account it appears that plants have been obtained " Bull. Am. Mus. Nat. Hist., vol. 23< 1907. P- 845. •' The possible exceptions are a single species of Corbicida reported by Whit- field from the upper portion of the "Hell Creek beds, " this genus being found in both brackish and fresh waters, and Ostrca glabra reported from the sec- tion near Yule, North Dakota. HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 219 at over fifty localities, and when these are brought together in a single list we have no less than 193 forms that have thus far been found in the lower member of the Fort Union. After eliminating the new forms, and those not specifically named, we have the following 84 species that are positively identified:^* a List of plants identified in lower Fort Union. *AraIia not at a Lesq. *Berchcmia midtinervis (Al. Br.) Heer. *Carpolithes lineatus Newb. *Celastrus alnifoliiis Ward. ^Cclastrus airvinervis Ward. *Celastn' Bull. Am. Mus. Xat. Hist., Vol. XXIII, 1907, p. 845. »=^ Newcastle folio (No. 107), Geol. Atlas U. S., U. S. Geol. Survey, 1904. 246 STANTON formation and there are indications that the hne of lithologic change is not a constant horizon throughout; thus it may be that some of the lower members represented as Laramie formation in this region be- long to the Fox Hills in other districts. According to Darton the "Laramie formation" (i.e., "Ceratops beds") of the Newcastle quadrangle consists of soft, massive sandstones intercalated with carbonaceous sandy clays. The thickness is 700 to 800 feet, as nearly as can be ascertained Next above the Fox Hills beds are 80 feet of gray clays overlain by sandstones. The sandstones consist mainly of fine-grained, loosely cemented sand of light-buff color, often having a thickness of 40 feet. They contain very charac- teristic concretions of gray color and great variety of shape. The beds of shale which occur interbedded among the sandstones of the Laramie formation are usually of dark gray color and in places lignitic, but no coal deposits have been found in them in this region. If Barton's observations are correct the rapid northward thinning of the sandstones beneath the "Ceratops beds" isduenot to erosion of the upper members but to lateral replacement of the lower members by shale. Northward from Weston County, Wyoming, to Hell Creek on the Missouri River in eastern Montana the presence of the " Ceratops beds" is indicated by the occurrence of the characteristic dinosaurs at many localities, though it may not be possible to trace the formation continuously all the way. At these localities the sandstones repre- senting the Fox Hills are either thin or entirely lacking, the " Ceratops beds" often resting, as at Forsyth and Myers, Montana, directly on a shale that is usually called Pierre, though Meek and Hayden early recognized the fact that its fauna contains a mixture of Pierre and Fox Hills species. Hell Creek, Montana. — This locality, just south of the Missouri River in eastern Montana, has been made prominent by the collections and descriptions of Mr. Barnum Brown''^ of the American Museum of Natural History. The base of the section is "Pierre shale," with a total exposed thickness of nearly 200 feet, which has yielded a "The Hell Creek beds of the Upper Cretaceous of Montana. Bull. Am. Mus. Nat. Hist., Vol. XXIII, 1907, pp. 823-845. "CERATOPS BEDS OF WYOMING AND MONTANA 247 marine invertebrate fauna showing a blending of Pierre and Fox Hills elements. Above that are "soft, arenaceous shales and interstratified beds of clay usually with a capping of sandstone" varying in thickness from 40 to 80 feet. These shales and sandstones have also yielded a mairine fauna and are referred by Brown to the Fox Hills. The "Ceratops beds" to which the local name of "Hell Creek beds" has been given are about 300 feet thick, and according to Brown may be divided into an upper and a lower member. The former is composed of arenaceous clays alternating with sandstones and car- bonaceous matter. The lower member consists of massive sandstone. . This is the most constant of the series. Itunconform- ably overlies the Fox Hills, as shown near the Cook ranch on Crooked Creek; also on Hell Creek. This unconformity is erosional in charac- ter. The unconformity has not been further described, but it may be inferred that the base of the sandstone rests on an uneven surface. Whether such an irregular base is due to an important unconformity or represents a long interval of erosion must be determined by other evidence. Similar phenomena are very common in non-marine deposits such as the typical Fort Union where almost every massive sandstone has an irregular base. A striking example of this may be seen in the Fort Union formation of the Bull Mountains, two miles south of Buckey's, Montana, where the unconformity is so striking that it was supposed to mark the base of the formation until the flora collected beneath it also proved to belong to the Fort Union. The vertebrate fauna includes the dinosaur genera Triceratops, Trachodon, Tyrannosaurus and Ornithomimus, together with other reptiles, fishes, and a few small mammals, and is nearly identical with the fauna of the "Ceratops beds" of Converse County except that not so many species have been collected. The invertebrate fauna is remarkable for the great development and differentiation of Unios, several of which have not yet been found elsewhere. The list as identified by R. P. Whitfield is as follows: Utiio CBSopiformis Whitf. U. corbiculaides Whitf. U. pyramidellus Whitf. U. verrucas if onnis Whitf. *f/. browni Whitf. 248 STANTON U. retusoides Whitf . U. percorrngata Whitf. U. posthiplicata Whitf. ft/, aldriclii White f Z7. holmesianus White U. vetustus Meek ft/, cryptorhynchus White *£/. biesopoides Whitf. U. cylindricoides Whitf. *f/. letsoni Whitf. *f/. gibbosaides Whitf. *t/. pyramidatoides Whitf. U. subtrigonalis Whitf. Spharhim planum M. & H. Corbicula subelliptica M. & H. "f^Campelorna jnultilineaia M. & H. ^Campeloma vetida M. & H. Campelomu producta White ■\Viviparns plicapressus White ^Cassiopella turricula White *Thauniastus limnmformis M. & H. Bulimus rhomboideus M. & H. Species marked * occur in "Ceratops beds" of Converse County. Species marked t occur at Black Buttes. On the faunal and stratigraphic evidence Brown's correlation of the "Hell Creek beds" with the " Ceratops beds" of Converse County, Wyoming, is fully justified. The flora is relatively meagre and accord- ing to Knowlton shows closer relationship with the overlying Fort Union. Of 13 identified species 2 are recorded in the Laramie, 3 in the Denver, and 8 in Fort Union and later formations. Between the "Hell Creek beds" and the typical "yellow beds" of the Fort Union there are 100 feet of lignite-bearing strata in which no dinosaurs have been found and which were tentatively assigned to the Fort Union for this reason, though the few plants obtained from them are included in the census just given. Areas in North and South Dakota. — Along the little Missouri River in the southwest corner of North Dakota A. G. Leonard has studied a section similar to that on Hell Creek and his description of it is published by Knowlton." The base of the section is a marine Cretaceous shale referred to the Pierre which is overlain by 80 feet '* Proc, Wash. Acad. Sci., vol. .\i, Xo, j;, lyog, p. 201. "CERATOPS beds" OF WYOMING AND MONTANA 249 of unfossiliferous sandstone and sandy shale probably representing the Fox Hills. Resting unconformably on this member are about 500 to 600 feet of "somber beds" extending up to the "typical buff and light-gray Fort Union shales and sandstones which are exposed near Yule P. O." The unconformity is presumably expressed by an uneven surface as at Hell Creek. About 200 feet above this unconformity dinosaur bones were collected which have been identified by C. W. Gilmore as Triceratops horridus Marsh, and several species of plants referred by Doctor Knowlton to the Fort Union were obtained rom the same and higher portions of the ''somber beds. " Among the invertebrates submitted by Doctor Leonard for examination one lot from near Yule consists entirely of Ostrea suhtrigonalis E. & S., which I considered sufficient evidence of Cretaceous age and stated in my manuscript report that if there is any Laramie in the region it probably includes this oyster bed. In a recent letter Doctor Leonard makes the following statements concerning the various collections from the , somber beds:" The plants which I sent to Doctor Knowlton and the specimens of Ostrea suhtrigonalis were not collected from the same stratum and came from localities several miles apart. As near as I could deter- mine the plants which came from near Yule are from a bed somewhat lower than the layer containing the shells {Ostrea). A few of the plants, those from near Marmarth and the mouth of Bacon Creek, are from the same beds as those containing the dinosaurs. Before Doctor Knowlton identified the Fort Union leaves from there I was inclined to regard as Cretaceous all the somber beds below the typical light-colored Fort Union, particularly after your statement regarding the probable age of Ostrea suhtrigonalis. I believe with you that we are here dealing with beds near the line between Cretaceous and Tertiary and it is possible that the upper 200 feet or so of the "somber beds" containing thick coal beds, may be Fort Union, and the lower portion which is barren of coal and contains dinosaurs may be Cretaceous. But one difficulty with this view is that Ostrea was found in the upper portion and several leaves identified as Fort Union species were found in the lower portion carry- ing the bones. The fossils do not seem to tell the same story as to the age of the beds in question. The significance of these facts and especially the occurrence of the oyster bed will be discussed later. 250 STANTON The same dinosaur horizon occurs on the Missouri River a few miles north of the mouth of the Cannonball River, where in 1908 1 collected bones identified as Ceratopsia and Trachodon from soft sandstones and shales approximately 100 feet above the top of the Fox Hills sand- stone, which is here fossiliferous and has its most northern outcrop on the west side of the Missouri near old Fort Rice, about 25 miles south- east of Mandan, North Dakota. The fossil plants listed by Knowlton from my collection were obtained from beds a few feet above the highest observed dinosaur bones. There is doubtless a continuous area of dinosaur-bearing beds extending through northern South Dakota and southern North Dakota between the Alissouri and Little Missouri rivers. Unpub- lished records furnished by Prof. J. E. Todd, former State Geologist of South Dakota, show localities for Triceratops and also for a number of brackish-water invertebrates in the region of Grand and Moreau rivers. The shells include Anomia, Ostrea glabra, Corhicula suhel- li plica, C. cytherifonnis, and C.occidentalis. Dinosaurs and brackish- water shells were also obtained in this region by F. V. Hayden on one of his early expeditions. These beds immediately overlying the Fox Hills along the Missouri in North and South Dakota might fairly be regarded as the basal part of the Fort Union formation, since it is the first named formation above the Fox Hills in the section, if it were not for the fact that Meek and Hayden limited the type section of that formation to the exposures along the Missouri from Fort Union to Fort Clark and up the Yellowstone into Montana, while these lower beds in question occur along the Missovi below Fort Clark. When the Fort Union formation was named by Meek and Hayden'"' they said it occupies the whole country around Fort Union [near the mouth of the Yellowstone] extending north into the British possessions, to unknown distances, also southward to Fort Clark. Seen under the White River group on North Platte River above Fort Laramie. Also on west side Wind River Mountains. In later publications both Meek and Hayden treated these lower beds as probably distinct from the Fort Union. Thus Meek'^ says: '^ Proc. Acad. Nat. Sci. Phila., 1861, p. 433. '«U. S. Geol. Surv. Terr., Vol. IX, 1876, p. XLIX. "CERATOPS beds" OF WYOMING AND MONTANA 25 1 Doctor Leidy also described among Doctor Hayden's collections from the lowest beds of the Upper Missouri Lignites near Moreau and Grand rivers, Nebraska, very probably belonging to the horizon of the Judith River group," some vertebrate remains which have been considered Cretaceous types by Cope. Hayden's last general statement on the Laramie problem in the introduction to Lesquereux's "Tertiary Flora "^^ indicates similar views, as the following quotations will show: The physical conditions under which the sediments of the upper strata of the Fox Hills group were deposited indicated a gradual change, from deep, quiet marine seas to shallow waters, which became at length brackish and finally entirely fresh waters, during which the purely marine invertebrate fauna perished, a brackish and purely fresh- water fauna taking its place As we proceed southward and westward from the Missouri River, the brackish beds increase in thickness until along the fortieth parallel they become three thousand feet or more, indicating, so far as can be determined, no break in the sequence from the Fox Hills group to the purely fresh- water strata of the Wasatch group The facts as we understand them at the present time would seem to warrant this general division, viz: a marine series. Cretaceous; gradually passing up into a brackish- water series, Laramie; gradually passing up into a purely fresh-water series, Wasatch. It is also probable that the brack- ish-water beds on the upper Missouri must be correlated with the Laramie, and that the Wasatch group as now defined and the Fort Union group are identical as a whole, or in part at least. The plants which are recorded in this volume began their existence at the base of the Laramie group and continued through the entire series, brackish and fresh-water. Localities on Yellowstone River in Eastern Montana. — Near Glen- dive the erosion of a low anticline has exposed the Pierre shale and overlying rocks in which Barnum Brown records the presence of Triceratops and Trachodont dinosaurs. Leonard has published'^ a detailed section and a list of invertebrates collected in the upper part of the Pierre shale which here has elements of both Pierre and Fox Hills faunas. The sandstones immediately above the Pierre have yielded no marine fossils and apparently belong to the "Ceratops "The Judith River formation was then supposed to overlie the Fox Hills. "U. S. Geol. Surv. Terr., Vol. VII, 1878, pp. III-VII. "Bull. U. S. Geol. Survey, No, 316, pp, 195-198. 252 STANTON beds" rather than to the Fox Hills. No discordance has been observed here at their base. Leonard's section records 535 feet of shales and sandstones above the Pierre. Pierre shale is exposed in a similar uplift west of Forsyth. Doctor Knowlton and I examined this section in 1908 near Castle Butte about 7 miles northwest of Forsyth, where we collected the following inverte- brates: Inoceramus sagensis Owen? Trigonarca {Breviarca) exigua M. & H. Lcda (Yoldia) evansi M. & H. TJietys? circular is M. & H. Solemya? sp. Mactra gracilis M. & H. Corbidamella gregaria M. & H. Aniauropsis paludina'formis M. & H. Anchura nebrascensis M. & H. Fasciolaria {Piestochilus) cretacea M. & H. Bacidities compressus Say Scaphites nodosiis Owen Placenticeras whitfieldi Hyatt Placenticeras inter calare M. & H. Above the Pierre are yellowish cliff-forming sandstones with thinner beds of shale dipping about 5° east which form the mass of Castle Butte and other conspicuous hills in the neighborhood. There is an estimated thickness of about 100 feet of these sandstones beneath the rocks that form the bluffs at Forsyth where the beds are horizontal. The contact between the Pierre shale and the overlying sandstone is very well exposed and shows beneath the massive cliff forming sand- stone, a few feet of soft sandstone alternating with clay shale appar- ently forming a transition zone. Mr. A. C. Silberling informs me that at and near Castle Butte he has observed remains of Triceratops, Claosaurns [Trachodon?], mam- mals, turtles, and fishes, showing that the sandstones belong to the " Ccratops beds. " He also reports the occurrence of Triceratops four miles north of Rosebud on Horse Creek east of Forsyth and near Junction City and Custer. The marine Cretaceous extends in a belt several miles wide from Castle Butte across country along the line of the Chicago, Milwaukee and St. Paul Railway, to Musselshell River, and uj) the Yellowstone to a point one mile west of JNIyers station, which is about 30 miles west "CERATOPS beds" OF WYOMING AND MONTANA 253 of Forsyth. Here also the "Ccratops beds" rest directly on marine shales and are exposed in prominent wooded bluffs to a thickness of about 300 feet. The following detailed section measured September 21, 1908, may be of interest: Section i mile west of Myers, Montana. Feet. 1. Massive, light-gray sandstone capping one of the highest buttes southwest of Myers, with fragmentary remains of Ceratopsia, Trachodon, and Ornithomimus 40 2. Thin-bedded sandstones and sandy shales with occasional large calcareous concretions in which Splicerium, Goniohasis, Cam- peloma? and fragments of turtle shell were collected 20 3. Massive greenish gray rather soft sandstone with many brown indurated bands and masses and frequent irregular deposits of ' ' clay ball " conglomerate especially near the base 35 4. Shale and very soft gray sandstone 30 5. Soft sandstone with band of brown concretions and indurations near middle, and fragments of dinosaur bone and crocodile scute near base 25 6. Massive gray sandstone with a brown indurated band at top 18 7. Shale, carbonaceous in upper part 10 8. Very soft, argillaceous sandstone 5 g. Carbonaceous shale with fragments of dinosaur bones i i 10. Clay shale, more or less sandy, with several carbonaceous bands 40 11. Very soft argillaceous sandstone with an indurated band near the base 10 12. Shale, mostly carbonaceous 10 13. Massive gray sandstone with indurated bands and masses and Ceratopsia bones at base 40 14. Shales and soft sandstone passing laterally into massive sand- stone 20 15. Soft sandstones and shales 15 16. Sandy shales with thin bands of sandstone apparently forming a transition to the beds below 12 17. Soft gray Pierre shale with a band of calcareous concretions in the upper part which yielded Baculites ovatus, Protocardia suh- quadrata to flood plain of Yellowstone River 75 Total 4o6i Westward from Myers to the neighborhood of Custer and Junction City the rocks lie in a broad shallow syncline with scarcely perceptible dips, so that the "Cera tops beds" are continuously exposed along both sides of Yellowstone River, and on the left bank they extend still farther west to Pompey's Pillar and beyond. The coal bed on the high land 4^ miles southeast of Bighorn station from which 254 STANTON Pepperberg and Barnett collected the plants listed by Knowlton is in the highest rocks exposed in this syncline and may belong to the typical Fort Union. At any rate it is higher than any of the dinosaurs of the Triceratops fauna which are reported to be common in the sand- stones immediately overlying the marine shale throughout this area. The " Ceratops beds" of this section are apparently directly connected on the northwest with the similar formation tentatively referred to the Laramie (?) which underlies the coal-bearing Fort Union of the Bull Mountains field, while on the south they follow approximately parallel with and usually east of the Chicago, Burlington, and Quincy Railroad to a connection with the lower part of the Piney formation in the Sheridan, Wyoming, area east of the Bighorn Mountains. Throughout these areas the most prominent feature of the formation consists of the exposures of light-gray, rather soft sandstones. Bull Mountains, Montana. — A preliminary account of part of this area with a description of the coal-bearing portion of the section has been published by L. H. Woolsey.^ He says of the Fort Union for- mation : it is composed chiefly of gray to buff sandstone alternating with gray shale. The sandstone, though extremely variable, is commonly massive and evenly distributed throughout the section. The base of the formation is strongly marked by contrast with a band of olive- green clay shale, which belongs to the next lower formation. This shale is well exposed along the Billings — Roundup road, 2 or 3 miles south of Buckey post-office, and may be traced westward across vari- ous branches of Razor Creek, through Pratt's ranch, and down Dean Creek to Musselshell River. In Woolsey's columnar section,^* under the term ''Beds on Dean Creek," it is represented as about 200 feet thick and as resting on the "Laramie" sandstones. The top of this shale is marked by the apparently eroded surface mentioned on p. 63. It has yielded a considerable number of fossil plants which Doctor Knowlton has referred to the Fort Union as follows: Plalanus hasilohata Ward Plalanus guillclmoe Gopp. Platanus raynoldsii Newb. •"Bull U. S. Geol. Survey, No. 341, 1908, pp 60-75 «0p. cit., PI. IV. "CERATOPS beds" OF WYOMING AND MONTANA 255 Popiilus cuneata Newb. Leguminosites arachioides Lesq. Onoclea sensibilis fossilis Newb. Sequoia nordenskioldi Heer Taxodium occidentale Newb. Glyptostrobus europceus Car piles sp. Populus rotundifolia Newb. Populiis nervosa elongata Newb. Populus amhlyrhyncha Ward A r alia notata Newb. Corylus rostrata Aiton. Viburnum sp. The "Laramie" and older rocks of the section are not described in Woolsey's report, but from the exposures I visited with Mr. R. W. Richards on Dean Creek and from others I have seen north of Mussel- shell Post-office it is evident that the "Laramie" includes the"Cera- tops beds" and that it is the same formation that is described by R. W. Stone^^ as Laramie on Fish Creek northeast of the Crazy Moun- tains, though there it and all the overlying formations are much thicker. Area northeast of Crazy Mountains, Montana. — The stratigraphy of the area just mentioned, which is about 50 miles west of the Bull Mountains, had been treated by Weed,^^ Douglass,^^ and Stanton and Hatcher,'^ prior to the work of Stone, above referred to, who made a detailed geologic map (not yet published) in connection with a study of the coal resources. After Weed's field studies were finished the work of Douglass brought out the fact that in this region a non-marine formation (now referred to the Judith River formation) is intercalated in the marine shales of the Montana group and that there is another coal-bearing 22 Bull. U. S. Geol. Survey, No. 341, 1908, pp. 78-80. 23 Weed, Walter Harvey: The' Laramie and the overlying Livingston for- mation in Montana, Bull. U. S. Geol. Survey, No. 105, 1893. The Fort Union formation, Am. Geologist, Vol. XVIII, 1896, pp. 201-21 1. Little Belt Mountains folio (No. 56), Geol. Atlas, U. S., 1896. ^* Douglass, Earl: A Cretaceous and lower Tertiary section in south cen- tral Montana, Proc. Am. Philos. Soc, Vol. XLI, 1903, pp. 207-224. A geological reconnaissance in North Dakota, Montana, and Idaho; with notes on Mesozoic and Cenozoic geology, Annals Carnegie Mus., Vol. V, 1909, pp. 266-278. 2^ Stanton, T. W., and Hatcher, J. B.: Geology and Paleontology of the Judith River beds. Bull. U. S. Geological Survey, No. 257, 1905, pp. 59, 60. Proc. Wash. Acad. Sci., August, 1909. 256 STANTON sandstone (now identified with the Eagle sandstone) at the base of that group- The shales and sandstones between these two have been named the Claggett formation, and the shale above the Judith River has been named Bearpaw. The latter is the equivalent of the upper part of the "Pierre" shale as developed at Glendive and elsewhere in eastern Montana and probably includes the equivalent of at least part of the Fox Hills, though the upper part of the Fox Hills is possibly represented in the base of the overlying "Laramie." The succession in the marine Cretaceous with Stone's measurements of the dififerent formations is as follows: Montana group: Feet. Bearpaw shale 700 to 11 00 Judith River formation 400 to 800 Claggett formation 400 to 800 Eagle sandstone 100 to 250 Colorado shale 1300 This stratigraphic succession is developed in most of the Cretaceous areas in Montana west of a line drawn across the state through the Pryor Mountains and the mouth of Musselshell River. For detailed descriptions of these formations the reader is referred to the papers cited — especially to the one by Stone for local details — but there are some facts that deserve mention in this connection. One of these important facts is that the Judith River formation with a vertebrate fauna of Trachodon, Ceratopsia, turtles, etc., occurs here in unquestionable stratigraphic sequence beneath about a thous- and feet of marine Cretaceous strata which in turn are overlain by beds containing a related dinosaur fauna. Not only is this true, but Douglass^® has recorded that the intervening Bearpaw shale [its lower part] has yielded specimens of Trachodon and other land dino- saurs directly associated with marine invertebrates, showing that these dinosaurs inhabited the neighboring land continuously from Judith River time until after the end of the Bearpaw. Brown records-^ a similar discovery of a dinosaur in the upper part of the Pierre shale of the Hell Creek region. ^* Proc. Am. Philos. Soc, Vol. XLI, p. 212. Annals Carnegie Mus., Vol. V, p. 276. "Bull. Am. Mus. Nat. Hist., Vol. XXIII, 1907, p. 826. CERATOPS beds' OF WYOMING AND MONTANA 257 Another important fact, brought out by Stone's work, is that the "Laramie" of Weed's mapping east and north of the Crazy Moun- tains in the Little Belt Mountain folio is really Eagle sandstone and the "Livingston" of the same area includes in its lower part the Claggett, Judith River, and Bearpaw formations. The Livingston formation was described as resting unconformably on the Laramie and older rocks and as composed largely of andesitic material, both of which features played a prominent part in correlating the Livingston with the Denver formation. It became necessary therefore to study the lithologic character of the various formations of the Montana group in this area, especially since certain beds in the Claggett and Judith River were observed to have an igneous appearance. On my request thin sections have been made of a number of specimens collected by Mr. Stone and those have been examined by Messrs. Johannsen, Calkins, and Stone. The specimens from the Eagle and a few of those from the Claggett and Judith River proved to be sand- stone without admixture of igneous material, but many others from both Claggett and Judith River and some from the Bearpaw are identified as tufaceous rock and contain much andesite, as the follow- ing extracts from the report will show: C. M. 71. Tuff. From the Claggett formation in T. 6 N., R. 15 E., sec. 14. Consists chiefly of grains of plagioclase and altered lava, and is probably a water-laid tuff. C. M. 74. Tufaceous rock. From near the top of the Claggett formation, 650 ft. above the Eagle sandstone, in T. 6 N., R. 15 E., sec. 13. Consists of angular fragments of plagioclase and andesite, and so far as seen contains no augite and no quartz. Probably is tufaceous. C. M. 75. Tufaceous rock. From the top of the Claggett formation, in T. 8 N., R. 12 E., sec. 29. Shows two pebbles of andesite in a matrix like No. 15. Probably is tufaceous. C. M. 76. Tufaceous rock. From near the base of the Judith River for- mation, in T. 6 N., R. 15 E., sec. 13. This rock is a conglom- erate with subangular to rounded pebbles, chiefly of altered porphyritic andesite and one pebble of fine sandstone with angular grains of quartz. The specimen may be an impure tuff. C. M. 78. Tufaceous rock. From near the base of the Bearpaw, at the mouth of Lost Horse Creek, T. 7 N., R. 11 E., sec. 17. Con- sists of angular grains of feldspar, lava, augite, magnetite, and quartz, with considerable chlorite. It is distinctly a tufaceous rock. 258 STANTON C. M. 79. Tufaceous conglomerate. From near the base of the Bearpaw formation, just below No. 78, at the mouth of Lost Horse Creek, T. 7 N., R. 11 E., sec. 17. This conglomerate consists of rounded pebbles of augite-andesite, or basalt, all of which are very similar to one another. The rock is probably a water- rolled tuff. These were selected from the report because they come from the immediate neighborhood in which plants, invertebrates, and verte- brates were obtained from the higher formations and there is no doubt as to the identity of the different horizons. Farther northwest near Summit and Dorsey and along the west side of the Crazy Mountains andesite tuffs are more common, but the marine Cretaceous forma- tions are there not so certainly identified. Only the following wn'll be cited: C. M. 14. Trachyte porphyry from Claggett beds 900 feet below fossils (609) determined as Bearpaw. Sec. 4, T. 7 N., R. 9 E. C. M. 15. Andesite tuff. From near the base of the Bearpaw; T. 7 N., R. 9 E., sec. 5. Bearpaw fossils collected above. Specimen consists of many feldspar and augite crystals, with some mag- netite and rock fragments with the texture of an andesite groundmass. The rock undoubtedly is essentially a pyroxene andesite tuff. The marine Cretaceous fossils collected above these specimens are: Ostrea pellucida M. & H. Avicula liuguiformis E. & S. Avicula nchrascana E. & S. Inoceramus cripsi var. harabini Morton Callista deweyi M. & H. Baculites compresses Say Brackish-water shells were reported by Weed^® "as occurring northeast of the Crazy Mountains in a bed of limestone, interbedded with the Livingston sandstones and conglomerates, a few hundred feet above the base of the formation." The locality as recorded on the field label is "Muddy Creek west of Gordon Butte and Cotton- wood Creek" and the species identified are: Ostrea subtrigonalis E. & S. Corbicula cyl/ieri/ormis M. & H. Corhula subtrigonalis M. & H. Corbula subtrigonalis var. perundata M. & H. ** Bull. U. S. Geol. Survey, No. 105, p. 33. "CERATOPS beds' OF WYOMING AND MONTANA 259 These fossils were referred by me to the Laramie in 1893, but at that time the Judith River formation was not recognized as distinct from the Laramie. They actually occur at this place in the Claggett with both Judith River and Bearpaw exposed above them in the same section as was observed by R. W. Stone, C. A. Fisher, and myself in 1907. The associated conglomerate mentioned by Weed is mostly composed of igneous pebbles but there is no suggestion of an uncon- formity beneath it nor at any other plane within the limits of marine sedimentation. With the facts above recited in mind the whole question of the age and relations of the Livingston formation is reopened, and inciden- tally the occurrence of 5 per cent of andesite in the Hell Creek beds recorded by Brown^® loses much of the significance that has been attached to it. Whatever may be true of the Livingston in the type area near the town of that name, the rocks assigned to it by Weed east of the Crazy Mountains, notably in the Lebo Creek section, certainly belong to several distinct formations ranging in age from well down in the Cretaceous to the lower Eocene. The rocks immediately overlying the Bearpaw shale in the Fish Creek section are thus described by Stone :^° The red and greenish sandstones overlying the Bearpaw shale con- stitute a distinct lithologic unit, ranging in thickness from 200 to 460 feet, and although containing some shale members, are sufficiently sandy and hard to form a conspicuous ridge. This formation is par- ticularly well exposed as a ridge along the western side of T. 5 N., R. 17 E., and crosses Fish Creek at the George Moore ranch. A mile farther north it swings to the west and can be traced readily all the way to Lennep. A peculiarity of this formation is its general reddish color and the occurrence of a layer of red, sandy, cannonball nodules near the middle. Its age has not been definitely determined, but lithologically it belongs to the overlying formation and hence it will be provisionally regarded as a part of the Laramie. The red color of this member has not impressed other observers. To my eye the prevailing tints are grays and browns, but the topo- graphic importance of the horizon has been recognized by all. The 2»Bull. Am. Mus. Nat. Hist., Vol. XXIII, 1907, p. 832. '" Op. cit., pp. 79, 80. 26o STANTON rocks are designated as "Fox Hills(?) " in Douglass' latest paper, but in his first description of this section that term was made to include the lower members which he now refers to the Laramie and which "contain in almost every good exposure bones of dinosaurs, such as Triceratops and probably Trachodon. " Some additional paleontologic evidence has been obtained from beds beneath the horizon of these dinosaurs. About 2^ miles northwest of George Moore's (formerly McClatchey's) ranch in sec. 23, T. 6 N., R. 16 E., the following suc- cession was observed in ascending order: Feet. 1. Valley underlain by Bearpaw shale not well exposed. 2. Light colored, friable, rather heavy bedded sandstone with a fossiliferous band at the top yielding Ostrea glabra, Anomia, Corbicula, Melania wyomingensis? etc 30 to 40 3. Dark shales and sandstones with nodules containing Ntictda planiniarginata, Leda (Yoldia?) evansi, Macira warrenana, Thracia {?), Lunatia, and other marine Cretaceous fossils. . 100 4. Light-colored sandstone 30 5. Soft sandstones and shales with bands of hard brown sandstone. One of these at the top contains many fossil plants 50 The plants were examined by F. H. Knowlton in 1903 and were then considered possibly older than Montana, though the evidence was unsatisfactory. Doctor Kjiowlton has recently again examined these plants and has pronounced them Livingston.^^ This plant horizon is on the top of the ridge formed by the rocks of Douglass' "Fox Hills (?) " which here dips 23° southerly. Some- what more than a mile along the strike in section 16 of the same town- ship bones and teeth of Triceratops and other dinosaurs were collected at the southern base of the ridge not more than 150 or 200 feet above the plant bed. A few fresh-water invertebrates have been collected from beds near the dinosaur horizon. These include Unio hrachyopistlms White Unio reliisoides Whitfield? Unio sp. Campcloma? sp. Lhnnaa? sp. " Proc. Wash. Acad. Sci., Vol. XI, No. 3, 1909, p. 194, "CERATOPS beds" OF WYOMING AND MONTANA 261 The Unios indicate close relationship with the fauna of the " Cera- tops beds" of Hell Creek and Converse County. Stone describes the upward stratigraphic succession as follows: Taken as a whole, the upper part of the Laramie formation is distinguished from the formations above and below by its light-gray color in comparison with their somber hues. It is composed largely of soft gray sandstone and variegated shale. The gray beds, from 1000 to 2400 feet thick, make a conspicuous valley across the middle of T. 6 N., Rs. 13 to 16 E. As a whole the formation weathers so readily that it normally forms low country, and for some miles in this area it coincides with the valley of Fish Creek. The gray beds of the Laramie formation are overlain, possibly with unconformity, by somber-colored sandstone and shale which may represent the Livingston formation. Sufficient paleontologic evidence has not been obtained, however,to determine the limits of these stratigraphic units. A section measured by C. A. Fisher and T. W. Stanton on a fork of Big Elk Creek gives a thickness of 5592 feet from the base of the Laramie to the base of the sandstone and grit of probable Fort Union age, and of 10,324 feet for the beds above the Bearpaw shale. Further field work is necessary before the lithologic and paleontologic dis- tinctions of the Laramie and Livingston formations can be deter- mined. The Fort Union formation, of unknown thickness, but exceeding 4,300 feet, is the youngest in this area. It is composed largely of sandstone, alternating with shale. The base of the formation is a particularly massive, coarse-grained sandstone which forms pro- nounced wooded ridges. Fort Union beds underlie the southern part of T. 6 N., Rs. 12 to 16 E., and compose the north and east bases, at least, of the Crazy Mountains. It was my privilege in 1908 under the guidance of Mr. A. C. Silber- ling to review the upper part of the Fish Creek section, from the Tri- ceratops horizon upward, especially in townships 5 and 6 north, ranges 15 and 16 E. The area is important because it was here that Douglass^^ found the first specimens of the primitive mammalian fauna on which the correlation of the Fort Union with the Puerco and Torrejon for- mations is based, and with one exception all the "Fort Union" mam- mals known were found within a few miles of this place. The strati- 32 Proc. Am. Philos. Soc, Vol. XLI, 1902, pp. 217-224. Annals Carnegie Mus., Vol. V, 1908, pp. 11-26. 262 STANTON graphic relations of the various paleontologic zones are easily deter- mined. In T. 6 N., R. 16 E., the gray beds forming the "upper part of the Laramie" of Stone's classification have at least the maximum thick- ness he assigns to them (2400 feet) as they underlie a belt considerably more than a mile wide with a dip of 23°. Except near the base where dinosaur remains are common, as already noted, fossils are rare in this part of the section. In the upper half a few localities farther west in the general region have yielded two or three species of Unio, a Viviparus, and Campeloma muUilineata, but nothing distinctive. Dinosaurs may range to the top of these "gray beds" but all the evi- dence now available on their upper limit are in the following excerpts from letters of Mr. Silberling: Dinosaur bones in top of Fox Hills and in base of Laramie [the thickness of which be estimates to be 3000 feet]. Dinosaur bone found in 1902 by myself 1800 or 2000 feet above base of [Laramie] beds, but there is some doubt as to the bones being in position. Dinosaur bones found in December 1908 within 600 feet of the top of the formation by A. Strand. I have not as yet seen the locality. Under date of January 28, 1909, Mr. Silberling writes: Don't be surprised if I should drop you a line some of these days and tell you John Winsbrough had found dinosaur bones in my Fort Union No. I. He has found a big limb bone over in T. 4 N., R. 16 E., some- where, but cannot at present locate it. Silberling's "Laramie" has the same limits as those given in Stone's description, and his "Fort Union" begins with the "somber-colored sandstone and shale which may represent the Livingston" according to Stone. Mr. Silberling was employed in this area for several months last year under my direction in collecting vertebrate and other fossils for the U. S. Geological Survey and the U. S. National Museum. He obtained an excellent collection of primitive mammals now being studied by Mr. J. W. Gidley, who states that they are related to the Puerco and Torrejon faunas of New Mexico with rather closer resem- blance to the Torrejon. In order to indicate the relative positions of his collections Silberling made use of prominent lithologic and topographic features to separate the rocks into three members which "CERATOPS beds" OF WYOMING AND MONTANA 263 he designated as "Fort Union Nos. i, 2, and 3," respectively. The lower member, or "No. i," and the middle member, or "No. 2," together constitute the softer, darker colored shales and sandstones with a combined thickness of about 1300 feet, while the upper member, or "No. 3," consisting of massive sandstones interbedded with shales, is more than 4000 feet thick and is identical with the whole of the Fort Union as recognized by Stone in this section and earlier by Weed in the Lebo Creek section. The lower and middle members were measured by Mr. Silberling and myself in sec. 15, T. 6 N., R. 15 E., where the rocks are well exposed, including the contacts with the "Laramie" gray sandstones below and with the massive basal sandstone of "No. 3" above. On account of the importance of the paleontologic collections obtained in these members in adjacent townships where the succession is identical it may be well to give the details. Section on branch of Fish Creek, Montana, in sec. 15, T. 6 N., R. 1 5 E. "Fort Union No. 3." a. Massive sandstone forming base of this member. "Fort Union No. 2." Feet. h. Shale with a few thin bands of sandstone 112 c. Gray sandstone 5 d. Shale _. 18 e. Gray limestone, weathering brown 3 /. Greenish gray shale with much soft sandstone of same color and brown ironstone concretions in lower third ; a few thin bands of sandstone in upper part and several concretionary layers near top 7°° 838 "Fort Union No. i." g. Brown, thinly cross-bedded sandstone forming top of a con- spicuous broken ridge 32 h. Shale 200 i. Greenish gray sandstone 15 j. Greenish gray shale 249 496 Total thickness of Nos. i and 2 1334 "Laramie." k. Rather soft, gray sandstone forming top of underlying formation. The lowest fossil plants listed by Knowlton as Fort Union came from a bed about 50 feet above the base of Silberling's lower member 264 STANTON of the Fort Union. A few fragmentary mammals remains, among which Mr. Gidley has identified ClcBnodon sp. nov., and MioclcEnus? sp., were obtained from about the same horizon which, it should be remembered, is above the highest recorded occurrence of dinosaurs in the region, and fully 2000 feet above the highest horizon where dinosaurs are abundant and positively known to occur. The most important collections of mammals were found in the middle member ("Fort Union No. 2"), and especially in the quarry opened in a thin sandstone 65 feet below the top of the member on the east side of Bear Butte in sec. 4, T. 5 N., R. 16 E. On preliminary examination Mr. Gidley has recognized Ptilodus montanus Douglass, P. gracilis Gidley, P. serratus? (Marsh), P. formosus? (Marsh), Mioclanus sp. nov., Euprotogonia sp. nov., Didymictis sp.nov., Ectoconus? sp. nov., Deltatherium? sp., Psittacotherium sp., Chriacus sp., Pantolamhda sp., Oxyclanus sp., Tricentes? sp., Pentacodon? sp., Mixodedes? sp. The two doubtfully determined species of Ptilodus described by Marsh occur in the " Ceratopsbeds" of Converse County, Wyoming. All the others are related to species in the Torrejon fauna of New Mexico, which was originally included in the Puerco. Clceno- donferox (Cope) and Pantolamhda cavirictiis? Cope, belonging to the same general fauna, were obtained in Silberling's upper and thicker member of the Fort Union, the sandstones of which yield abundant remains of a typical Fort Union flora. The upper and middle mem- bers also yielded the reptiles Champsosaurus puercensis? Cope and Crocodilus sp. In both the middle and the upper members fresh-water shells are common at several horizons. They include Campeloma multilineata, which in some of its varieties ranges from the Laramie of the Denver basin and the "Ceratops beds" into the Fort Union, Viviparus, and several species of Unio, most of which appear to be undescribed. The Unios are all of simple types and do not include any of the pecul- iarly sculptured forms like those of Hell Creek, Converse County, and Black Buttes. As compared with the Bull Mountains section the "Laramie" of Fish Creek is much thicker, the lower and middle members of the Fort Union (1300 feet thick) correspond lithologically and strati- graphically with the 200 feet of shale on Dean Creek, and the upper member, though much thicker, resembles the coal-bearing Fort Union "CERATOPS beds" OF WYOMING AND MONTANA 265 of the Bull Mountains. If these resemblances could be accepted as identity in each case the smaller thicknesses in the Bull Mountains could be explained as due cither to slower deposition, owing to greater distance from the source of materials, or to partial removal by erosion during one or more periods. From eastern Montana westward to the Crazy Mountains there appears to be a progressive thickening of the beds lying between the marine Cretaceous and what has hither- to been generally called the typical Fort Union, and there is a similar thickening southward from Hell Creek to Converse County, Wyom- ing, but in all this region the lowest bed containing the Triceratops fauna is never very far above the same general horizon of the marine Cretaceous, and the evidence for an important period of extensive erosion at its base is very slight — not greater than the evi- dence for such erosion at the base of the typical, or "upper," Fort Union. Area east of the Bighorn Mountains near Sheridan, Wyoming. — The formations of this area have been named and described by N. H. Darton." The coal-bearing rocks have also been studied by J. A. Taff.^^ The highest recognized marine Cretaceous is the Parkman sandstone with a fauna related on the one hand to that of the Claggett and on the other to the Fox Hills fauna. Evidence was obtained by C. A. Fisher and myself in neighboring areas on the north that there are marine shales above the Parkman sandstone, but in this area the beds are not all well exposed and the details of the stratigraphy — especially the contacts between some of the formations — have not been worked out. Above the Parkman sandstone is the Piney formation, consisting of dark shales with coaly seams alternating with beds of massive light-colored sandstone. In .the Dayton quadrangle according to Darton its thickness is 2000 to 3000 feet and it is believed to be entirely of fresh-water origin. At a locality one-half mile north of Parkman, Wyoming, a sandstone in the Piney about 1000 feet above the Park- man sandstone has yielded some fragments of dinosaur bones together with imperfectly preserved species of Unio and other fresh-water shells '^ Bald Mountain-Dayton and Cloud Peak- Fort McKinney folios (Nos. 141 and 142), Geol. Atlas U. S., 1906. Geology of the Bighorn Mountains, Prof. Paper, U. S. Geol. Survey, No. 51, 1906. ^* Bull. U. S. Geol. Survey, No. 341, pp. 123-150, 1909. 266 STANTON which, while not specifically identifiable, are apparently of types that elsewhere occur in the " Ceratops beds" and older formations. This is the only locality in the region where dinosaurs have been found.^ The next higher formation in Barton's classification of the Dayton quadrangle is the De Smet, which in its upper half contains the principal coal beds mined at Monarch, Carneyville, Dietz, and else- where in the region. The lower member beneath these coals is described by Taff as "essentially all shale or ... . shaly in character and prevailingly dull drab, bluish and brown in color, " and has an estimated thickness of 2500 to 2800 feet. The upper member, of approximately the same thickness in the area studied by Taff, is lighter-colored and contains a larger proportion of sandstone and coal. The fossil plants of both members belong to the Fort Union flora according to Knowlton, and the invertebrates which have been found only in the upper member are also Fort Union forms. Attention is called to the fact that all the plants from this district listed by Knowlton as lower Fort Union, except one small lot from near Buffalo, which will be referred to again, are from the De Smet forma- tion and the lowest of them are from several hundred feet above its base. In other words they have nothing whatever to do with the "Ceratops beds" which are without much doubt represented in the Piney formation. Farther south in the Fort McKinney quadrangle and in the southern part of the Dayton quadrangle the Kingsbury conglomerate has been mapped between the Piney and De Smet formations, though Darton^^ states that it probably develops out of the lower part of the De Smet. Taff's work in tracing individual coal beds and associated rocks led him to believe that the Kingsbury conglomerate is con- siderably higher in the section. His observations are recorded^^ as follows : A considerable part of the coal-bearing rocks change in character in certain respects toward the south, in the general direction of the strike of the beds. Near the State line, in T. 58 N., R. 86 W., cer- ** Hatcher has reported horned dinosaurs east of the Bighorn Mountains about 40 miles south of Buffalo, Wyoming, but nothing is known as to their stratigraphic position. ''Bald Mountain, Dayton folio (No. 141) Geol., Atlas U. S., 1906, p. 8. " Bull. U. S. Geol. Survey, No. 341, p. 131. "CERATOPS beds" OF WYOMING AND_MONTANA 267 tain light-colored sandstone strata in the upi:)er part of the lower member thin out and disappear southward along the strike. Some brownish and yellow sandy strata that lie still higher, near the top of the same member on Tongue River, seem to thin out toward the south and give place to dull-colored shale or sandy strata on Goose and Beaver creeks in T. 55 N., R. 85 W. These conditions seem to indicate that the parting between the two members rises in the rock section toward the south. The upper member also changes in character southward along the strike of the rocks. From the central part of the field southward the differentiation between the sandstone and the shale strata becomes less distinct. The sandstone on the whole is duller in color, and near the southern boundary of the mapped area the sandstone beds contain pebbles of limestone, quartz, and chert. In the southeastern part of T. 54 N., R. 83 W., and in T. 54 N., R. 84 W., many hundred feet of strata in the central part of the upper member merge into conglomer- ate. The constituent parts of the conglomerate become coarser rather abruptly on the approach to the Paleozoic rocks of the Bighorn Mountains, upon which the conglomerates overlap unconformably. The exposed section of conglomerate strata is more than 1000 feet thick between Little Goose and Sandy Creek valleys, at the base of the Big- horn Mountains, on the southern border of the Sheridan field. The gradation from the conglomerate into the sandy and shaly strata takes place toward the east and north, and involves almost the whole section of this member from the Tongue River coal group upward nearly to the top of the rock section. The economic bearing of the conglomerate is a negative one, for the coal beds thin out and disappear near its outer fringe. The Kingsbury conglomerate has a maximum thickness of at least 2500 feet west and southwest of Buffalo, Wyoming, where the basal beds and all the underlying formations are sharply upturned. Some of the beds are very coarse and the boulders and pebbles include granite and all of the harder Paleozoic rocks of the Bighorn Mountains. Fossil plants and invertebrates were collected here by Mr. T. E.Williard in 1907 and additional collections, including a mammal jaw found by Mr. Gale, were obtained by Messrs. H. S. Gale, C. A. Fisher, and myself in 1908. Mr. Gale made a plane table map (unpublished) and structure section of several land sections in the southwest part of T. 50 N., R. 82 W., and in the northwest corner of T. 49 N., R. 82 W., which enabled him to determine accurately the stratigraphic positions of the various collections and to demonstrate that there was a large 268 STANTON amount of erosion of the underlying beds before the conglomerate was deposited. Taking as a datum plane a fossiliferous band in the Parkman sandstone it is seen that the beds between it and the base of the conglomerate vary in thickness from about 1300 feet to more than 3000 feet in a distance of two miles along the strike. In a small collection of fossil plants from a white, friable sandstone about 2000 feet above the Parkman and nearly the same distance below the Kingsbury in sec. 6, T. 49 N., R. 82 W., Doctor Knowlton has identified 5 Fort Union species which are listed in his paper. What relation this horizon bears to the Piney farther north where dinosaur bones were found is still an open question. The most impor- tant collections, however, are from finer strata interbedded with the Kingsbury conglomerate about 600 feet above its base. From this part of the section the collections include a considerable list of Fort Union plants, a few land and fresh-water invertebrates most of which are undescribed but seem to be related to Wind River and Wasatch rather than to earlier forms, and a part of a small mammal jaw with the teeth in place. Concerning this jaw Mr. J. W. Gidley reports that it is identical with a genus and species occurring in the 'Silberling quarry" in Silberling's Fort Union No. 2. Apparently the genus and species are new, so that at present about all that can be said of the Gale speci- men is that it is closely allied to Tricentetes, a genus found in the Torrejon of New Mexico. The close correspondence of the specimen to like portions found in the bed of the Silberling .quarry, leaves little doubt that the horizons are equivalent. All the evidence both paleontologic and stratigraphic tends to prove that the unconformity at the base of the Kingsbury conglomerate is well up in the Fort Union. Since it has been proved to hold this high position the assertion has been made that the unconformity is purely local and relatively unimportant because the conglomerate extends for only 30 or 40 miles along the strike. That it is unimpor- tant and local may be true, but such an assumption should not be made until it has been compared with other observed unconformities which also may or may not be local. Bighorn Basin. — Except in the extreme northern part it has been difficult to recognize in Bighorn Basin the same subdixisions between "CERATOPS beds" OF WYOMING AND MONTANA 269 the Colorado shale and the Fort Union that have been used as for- mations in Montana, because both the Claggett and the Bearpaw as marine formations are either very thin or entirely represented by non- marine deposits. The geology has been described by Fisher^" who did not attempt to classify these higher rocks and brief abstracts of the more detailed work by Woodruff^^ and Washburne*" have been published. In the Shoshone River section below Cody there is a conglomerate which is comparable in many respects with the Kingsbury. The basal bed exposed at the mouth of Sage Creek contains many pebbles of igneous rock, abundant rounded fragments of silicified wood and many pebbles that look like the underlying Cretaceous rocks. Fort Union plants identified by Doctor Knowlton were obtained from beds in- tercalated in the conglomerates, as well as above them, and they range down some distance below their base. The presence of the "Ceratops beds" is indicated at several localities in the Basin by fragmentary dinosaur bones and by a few fresh-water invertebrates which are usually associated with them, though nothing strictly distinctive has been found among the invertebrates. In every case the dinosaurs were found in beds lower than the lowest recognized Fort Union plants. Similar stratigraphic relations are found just north of Bighorn Basin near Belfry and between Bridger and Red Lodge, Montana, except that the marine formations are recognizable and the Fort Union is much thicker and contains many workable coals. Dinosaur bones are found in the part of the section designated as "Laramie" in Woodruff's*' sketch and associated with them are the following inverte- brates: Unio sp. Related to U. brachyopisthus White U. pyramidatoides Whitfi.eld? U. cylindricoides Whitfield? U. verrucosiformis Whitf. SphcBrium Physa '* Fisher, C. A. : Geology and Water Resources of the Bighorn Basin, "Wyo- ming, Professional Paper, U. S. Geo!. Survey, No. 53, 1906. ^'Woodruff, E. G. : Coalfields of the southwest side of the Bighorn Basin, Wyoming, Bull. U. S. Geol. Survey, No. 341, p. 200-219. *" Washburne, C. W. : Coalfields of the northeast side of the Bighorn Basin, Wyoming, Idem., pp. 165-199. " Bull. U. S. Geol. Survey, No. 341, p. 96. 270 STANTON Columna Viviparus trochiformis M. & H. ? Campeloma muUilineata M. & H. Goniobasis tenuicarinata M. & H. The fauna is closely related to that of Hell Creek and Converse County. The lowest Fort Union plants recorded occur several hun- dred feet higher, except a fig identical with one found in Converse County and Hell Creek. Black Buttes, Wyoming. — This locality has been the subject of much discussion ever since 1872, when the dinosaur, Agathanmas sylvestris Cope, now referred to the Ceratopsia, was found there. The following description by Stanton and Knowlton*' will show the paleontologic characteristics of the dinosaur-bearing formation and the relative positions of the various species: The most prominent feature of the section at Black Buttes is the massive bed of sandstone, somewhat over 100 feet thick at the base of the exposure, forming steep hills and cliffs northeast of the railroad opposite the station and passing beneath the surface by its dip of 9 or 10 degrees near the coal mine. The upper portion of it is also exposed on the south side of Bitter Creek valley, about a mile from the station. All of the Laramie fossils, whether plants, invertebrates, or vertebrates, that have hitherto been described or listed as coming from Black Buttes were obtained from the overlying beds within about 100 feet of the top of this massive sandstone. The original specimen of Agathaumas sylvestris was found about 20 feet above it, and the plants that have been described came from the same horizon and from several higher bands up to the bed overlying the principal coal, some 60 or 75 feet higher. The invertebrates from this locality have about the same range. Most of the beds vary considerably in character and thickness within short distances, but the fossiliferous and overlying portions of the sections may be described in general terms as a series of variable sandstones, clays, and coal beds exposed in low hills and ridges with a dip of 9 or 10 degrees eastward at the base, but decreas- ing in the upper portions to 5 or 6 degrees, which is about the same as the dip of the overlying Wasatch beds. The character of the mollusks shows that the lower beds were mostly deposited in brackish waters, but that there were alternations of fresh waters in which the genus Unio thrived with an abundance of individ- uals and great variety of species, [and several fresh-water gasteropods were common. Between the top of the massive sandstone and the *^ Bull. Geol. Soc. America, Vol. VIII, 1897, pp. 143-145. CERATOPS beds' OF WYOMING AND MONTANA 27 1 Dinosaur horizon there is a band filled with brackish-water fossils, including Ostrea glabra var. arcuatilis Meek; Anomia micronema Meek; Corhula undifcra Meek, and Modiola sp. The greater num- ber of the Black Buttes invertebrates, however, have been obtained from strata some 40 or 50 feet higher, and consequently a little above the Dinosaur bed. Here there is a band which in some places is about four feet thick, almost wholly made up of shells. By far the most abundant species is Corhicula fracta Meek, and immediately associ- ated with it are Corhicula occidentalis M. & H.; Neritina haptista White; N. volvillneata White, and Melania wyomingensis Meek, all of which probably lived in slightly brackish water, for this species of Melania has almost invariably been found associated with brackish water or marine forms, although it is referred to a fresh-water genus. At the base of this shell bed and immediately above a coal seam Unto shells are abundant. These purely fresh-water forms are found on the slope mingled with the Corhicula shells, but all that were found in situ were either at the base of or a few feet above the Corhicula bed. The Unione fauna is strikingly like that of the Ceratops beds in Converse County, as the following list of species will show: Unio couesi White Unio propheticus White Unio aldrichi White Unio proavitus White Unio holme sianus White Unio endlichi White Unio crypiorhynchus White Unio hrachyopisthus White Unio goniambonaius White Unio dance M. & H. Immediately above the Corhicula bed a band is locally filled with Tulotoma thompsoni and occasional Unios, and it is overlain by shales containing Ostrea and Anomia in the lower part and the following species above: Unio couesi White Corbula undifera Meek Corbula subtrigonalis M. & H. Cassiopella turricula White Goniobasis gracilenta M. & H. ( ?) Viviparus plicapressus White Campeloma vettda M. & H. Campeloma multilineata M. & H. The fossil plants of Black Buttes, ranging through the same strata as the invertebrates and extending up a few feet higher to the bed just above the principal coal, are as follows: Proc, Wash Acad. Sci., August, 1909. 272 STANTON Apeibopsis discolor Lx. Cissites lobato-crenata Lx. Cissites tricuspidata Heer Ficus planicostata Lx. Ficus planicostata latifolia Lx. Grewiopsis saportanea Lx. Grewiopsis tenuifolia Lx. Myrica torreyi Lx. Ophegrapha antiqua Lx. Pisonia racemosa Lx. Podogonium atnericanum Lx. Quercus viburnijolia Lx. Rhamnus salicifolius Lx. Sapindus caudatus Lx. Sequoia acuminata Lx. Smilax grandifolia Ung. Viburnum marginatutn Lx. Viburnum platanoides Lx. Viburnum rotundifolium Lx. Viburnum whyinperi Heer Zizyphus fibrillosus Lx. The fresh-water element of the invertebrate fauna shows close relationship by means of specific identity with the fauna of Hell Creek and Converse County, while the brackish-water species are similarly connected on the one hand with the Laramie fauna of Crow Creek, Colorado, which is in the Denver Basin, and on the other with the Mesaverde" fauna of Point of Rocks and other localities in southern Wyoming and western Colorado. Of the fresh-water forms also Tulotoma thompsoni ranges down into the Mesaverde and a few of the Unios are represented by related species in that formation. While this range removes such species from the list of distinctive Laramie species it docs not impair their value as evidence of the Cretaceous age of the beds where they are found. The fossil plants of Black Buttes also show some species in com- mon with Converse County though according to Knowlton the rela- tionship is not close. The flora is apparently more closely connected with that of Point of Rocks, which is in the upper part of the Mesa- " In the current work of the Coal Division of the U. S. Geological Survey the formation names used in western Colorado have been applied in southern Wyoming. The Mesaverde corresponds in a general way with the combined Eagle, Claggett, and Judith River formations of Montana and the Lewis shale is the approximate equivalent of the Bcarpaw. "CERATOPS beds" OF WYOMING AND MONTANA 273 verde and separated from the Black Buttes horizon by the marine Lewis shale. Lesquereux" recognized the close relationship of the floras of these two localities, saying that they had 9 species in com- mon.^^ That the advance in science of the last 30 years has not changed this relationship is shown by two small collections obtained last year near Table Rock from the Black Buttes horizon in which the following species were identified by Doctor Knowlton and referred to "age probably same as at Black Buttes." ^Viburnum marginatum Lesq. V. contortum Lesq. *Myrica torreyi Lesq. Sequoia, reichenbachi (Gain.) Hear Sequoia heeri Lesq. ^Dryophyllum subfalcatum Lesq. *Ficus dahnatica Ett. Ficus planicostata Lesq. Ficus sp. *Phyllites triloba Kin. Of the nine identified species five (marked *) are reported from Point of Rocks. The general relations of these beds are clearly presented by A. R. Schultz*^ who calls them the Black Buttes coal group. He states that the Black Buttes coal group rests conformably on the Lewis shale, but that the overlying Black Rock coal group is separated from it by an unconformity of considerable magnitude. This unconformity was not observed by Doctor Knowlton and me in 1896, but last year, in company with Messrs. Veatch, Schultz, and Fisher, we visited a locality a few miles south of Black Buttes, where the true relations are strikingly apparent, as within a short distance the Black Rock coal group overlaps the Black Buttes coal group and rests directly on the Lewis shale. The beds above the unconformity contain a Fort Union flora. As evidence of the great geographic distribution of the Black "Tertiary Flora, U. S. Geol. Survey Terr., Vol. VII, 1878. *^ Doctor Knowlton admits only 8 species in common and says of them (Bull. U. S. Geol. Survey, No. 163, 1900, p. 77) "Several areopen tomoreor less question, as they depend on one or two fragments. " *' The northern part of the Rock Springs coalfield, Sweetwater County, Wyoming, Bull. U. S. Geol. Survey, No. 341, pp. 256-268. 274 STANTON Buttes invertebrate fauna at about the same horizon the following lists of fossils collected by Robert Forrester in the Laramie of south- west Colorado may be given : Beaver Creek, T. 34 N., R. 5 W., near boundary line between La Plata and Archuleta counties, Colorado. Coal measures immediately above the Lewis shale. Anomia sp. Related to A. micronema Meek. Modiola laticostata White. Corbicula sp. Related to C. suhelliptica M. & H. Corhicula occidenialis M. & H. Corbula undifera Meek Melania wyomingensis Meek? Yellow Jacket Creek, T. 34 N., R. 5 W., "Coal measures above Lewis shale." Ostrea sp. Unio holmesianus White Unio hrachyopisthns White Unio verrucosiformis Whitfield? Unio sp. Undescribed, possibly two species. Ttdotoma thompsoni White Campeloma? sp. Neritina sp. Mr. James H. Gardner, of the U. S. Geological Survey, who has done detailed work in the region, verifies the identification of the horizon at the localities where these collections were made. Area west of Rawlins, Wyoming. — At Black Buttes the marine Cre- taceous and immediately overlying rocks dip gently eastward passing under beds of Wasatch, Green River, and Bridger age in the broad syncline of the Great Divide Basin. They again come to the surface with a westward dip a few miles west of Rawlins and 60 to 70 miles east of Black Buttes. A summary description of the section exposed here is published by E. Eggleston Smith,''^ to which the reader is referred for lithologic and areal details. The paleontologic material dis- cussed below was collected in part by the field parties of Messrs. Smith and Ball and in part by myself. The estimates of thickness are Smith's. The Mesaverde formation, which is coal-bearing and about 3600 feet thick, is not very fossiliferous in this area, but a few marine inverte- brates have been obtained in the lower part and near the top there is *'' The eastern part of the Great Divide Basin coalfield, Wyoming, Bull. U. S. Geol. Survey, No. 341, pp. 220-242. "CERATOPS beds" OF WYOMING AND MONTANA 275 a thin bed of mostly brackish-water shells, among which the following have been identified from a locality in T. 22 N., R. 89 W.: Ostrea suhtrigonalis E. & S. Ostrea glabra M. & H. Anomia micronema Meek Mytilus subarcnaius M. & H. ? Corbicula occidentalis M. & H. Mclania wyomingensis Meek? Odontobasis buccinoides White Above the Mesaverde is the Lewis shale, about 1500 feet thick, with some layers of concretionary sandstone which are often fossilif- erous. In the township above-mentioned these yielded the following marine fossils: Ostrea sp. Modiola galpiniafia (E. & S.) ? Avicula nebrascana E. & S. Syncyclonema rigida H. & M. Nucula sp. Protocardia subquadrata E. & S. Corbula sp. Dentalium gracile H. & M. Anchura sp. Baculites ovatus Say Scaphites nodosus Owen The same fauna, with some additional species, such as Avicula fibrosa M. & H. and Lucina occidentalis (Morton) is also found in calcareous concretions in the dark shales. The coal-bearing formation, about 3900 feet thick, conformably overlying the Lewis, is called Laramie in Smith's report. Near the middle of the formation in T. 22 N., R. 89 W., in line across the strike from the Mesaverde and Lewis localities above recorded, Corbula suhtrigonalis M. & H. was collected and 10 feet higher Tulotoma thompsoni and Campeloma sp. occur. Fuller collections were obtained from this formation from the adjoining township on the south. Here a brackish- water bed about 1000 feet above the base yielded Ostrea glabra M. & H. Ostrea siibtrigonalis E. & S. ? Anomia micronema Meek Mytilus subarcuatus M. & H. ? 276 STANTON Corbicula cytheriformis M. & H. Corbicula fracta Meek PanopcBa simulatrix Whiteaves? Neritina volvilineata White This is evidently a recurrence with sHght modifications, of the brackish- water fauna found at the top of the Mesaverde and it is prac- tically identical with the brackish- water element of the fauna at Black Buttes. Considerably higher in the section near the middle of the ' 'Lara- mie" the following fresh-water shells were obtained: Unio goniamhonatus White Unio stantoni White Unio verriicosiformis Whitfield Unio hrachyopisthus White? Tulotoma thorn psoni White Goniohasis sp. All of these species except one are found at Black Buttes and that one occurs at Hell Creek and Converse County. The stratigraphic, areal, and structural relations and the faunal evidence all seem to justify the correlation of the "Laramie" of this area with the Black Buttes coal group. Above his Laramie Smith recognizes two unconformities between which there are coal-bearing strata with a total thickness estimated at 8780 feet which he treats as ** undififerentiated Tertiary." The lower part is conglomeratic, especially about 1000 feet above the base, where the conglomerate contains pebbles of both Paleozoic and Cre- taceous rocks. Fossil plants recognized as Fort Union by Knowlton have been obtained from several horizons in this thick series. Above the upper unconformity there are 900 to 1800 feet of coal- bearing beds referred by Smith to the Wasatch. The basal con- glomerate is in large part derived from granite with some sedimentary pebbles. The most important paleontologic evidence as to the age of this formation consists of some teeth of Coryphodon, a mammalian genus characteristic of the Wasatch, and an unstudied flora which is said to be different from the Fort Union. The fresh-water shells obtained also indicate Eocene later than Fort Union. Carbon County, Wyoming. — The area adjoining that mapped by E. E. Smith on the east and extending from Rawlins about 50 miles "CERATOPS beds" OF WYOMING AND MONTANA 277 east to Medicine Bow had been previously studied by A. C. Veatch/* and as Veatch at first had supervision of Smith's work the same strati- graphic units were recognized and used in both areas. With some variations in thickness and hthologic character the Mesaverde and the Lewis of the two areas correspond very closely. The " Laramie" of Smith's report was called "Lower Laramie" by Veatch and extends up to the same unconformity, with a reported thickness of 6500 feet. Veatch's "Upper Laramie" of this area with the addition at the top of 1200 feet of beds referred to Fort Union, apparently corresponds very well with Smith's "undifferentiated Tertiary." The " Upper Laramie" has a basal conglomerate " composed largely of pebbles derived from the underlying Cretaceous rocks" and in part of the area it overlaps several of the older formations. It includes the coal beds mined at Carbon, Dana, and Hanna. The flora associ- ated with the coal at Carbon has long been known and has been the subject of much discussion. It has usually been classed with the flora of Evanston as later than the other floras formerly referred to the Laramie, but the question whether it belongs to the Fort Union or to some other flora must wait until the plants have been thoroughly restudied with fuller collections made with careful reference to the stratigraphy of the region. The invertebrate collections from both the "Upper" and "Lower Laramie" have been very meager. The localities and horizons of the different lots as given on the field labels have been compared with the published geologic map, and the evidence will be recorded as it stands. A brackish-water fauna, represented by Osirea and Corbicula, was collected at several localities in the "Lower Laramie" and also in the Lewis and Mesaverde. Only one lot is from a locality mapped as "Upper Laramie" and that is from sec. 21, T. 24 N., R. 81 W., at the lower boundary of the formation where lower beds are probably exposed. As brackish-water shells have not been found above this major unconformity elsewhere in the region it is fair to assume that this lot was collected below it. A fresh-water fauna like that of Black Buttes is indicated by a " Coalfields of east-central Carbon County, Wyoming, Bull. U. S. Geol. Survey, No. 316, 1907, pp. 244-260. 278 STANTON number of small lots from the "Lower Laramie." The largest of these, from sec. 16, T. 24 N., R. 84 W., contains Unio couesi White Unio verrucosiformis Whitfield Unio sp. Campeloma muUilineata M. & H. Cassiopella turricida White At several other localities Tulotoma thompsoni White was collected with the other gastropods of this list. This fresh-water fauna was collected at but one locality mapped as "Upper Laramie." That is in sec. 10, T. 23 N., R. 80 W., near the contact where "Upper Laramie" is overlapping "Lower Laramie" and Lewis, making it probable that the collector was in error as to the line between the formations. Other localities in the "Upper Laramie" yielded Unio prisciis M. & H., Unio sp., Viviparus raynoldsanus M. & H., and Campeloma muUilineata, all of which occur in the Fort Union, though the last named species is also widely distributed in the Laramie and " Cera- tops beds." In this area Hatcher" has recorded the presence of " horned dino- saurs and Hadrosauridae on the North Platte River opposite the mouth of the Medicine Bow, about 35 miles below Fort Steele, Wyoming." This locality is mapped by Veatch as "Lower Laramie" and it is only about a mile from the place (sec. 16, T. 24 N., R. 84 W.) in the same formation where the fresh-water shells above listed were obtained. Mr. Veatch believes, however, that the dinosaurs were found in a bluff of "Upper Laramie" about a mile and a half above the mouth of Medicine Bow. His belief, it is said, is based on statements of a resident of the region, that large fossils were once collected there. The invertebrate fossils in the lower formation indicate to my mind that the dinosaurs ought to be where Hatcher said he found them, but the locality ought to be restudied.^" ^'Am. Naturalist, Vol. XXX, 1896, p. 118. '" This is perhaps an appropriate place to record another reported occur- rence of dinosaurs in a high horizon. Last year Mr. James H. Gardner collected bones referred to Triceratops, Trachodon, Tyrannosaurus, etc., near Ojo Alamo on Coal Creek, northwestern New Mexico, at a locality where Mr. Barnum Brown had previously found dinosaurs. Mr. Gardner believes that the bed from which he collected belongs to the Puerco and that it is several CERATOPS BEDS OF WYOMING AND MONTANA 279 So far as it goes the evidence in the Carbon County field tends to show tliat the "Lower Laramie" is conformable and continuous with the marine Cretaceous and that it includes the equivalents of the Black Buttes coal group and of the "Ceratops beds" of Converse County. It is also probable that the "Upper Laramie" is Fort Union. In thickness of strata and in the number and character of its coals this formation seems to be strictly comparable with the Fort Union of the Sheridan and Red Lodge fields. General Stratigraphic and Structural Relations It has been shown that, within the large area considered, the "Ceratops beds" with the Triceratops fauna are always pretty closely associated with the uppermost marine Cretaceous strata or are separated from them by transitional brackish-water beds. They are always overlain by a thick series of rocks containing a Fort Union flora in which no dinosaurs have been found, and in the Fish Creek, Montana, region this overlying series also contains primitive mammals related to those of the Puerco and Torrejon faunas. Throughout a large part of the area no evidence of an unconformity beneath the "Ceratops beds" has been found, while higher in the sec- tion unconformities have been demonstrated or suggested at a num- ber of places. Unconformities have been reported below tbe " Cera- tops beds" on Hell Creek, Montana, on the Little Missouri in North Dakota, and in Weston County, Wyoming, but in none of these cases has any proof been furnished that the erosion interval is important. The fact seems to be overlooked that in irregular non-marine deposits like these, which were in part at least fluviatile, an eroded surface should normally be expected at the base of a sandstone, but such an eroded surface may mean a time interval too brief to be worthy of con- sideration in geologic history. Because Ceratopsia were found in the Denver Basin above an unconformity that has been interpreted as representing an enormous time interval, is not a valid reason for hundred feet above the unconformity separating that formation from the Laramie. If that is true it is somewhat remarkable that all the fossils found here are reptilian and show considerable variety, while other Puerco localities not many miles away yield chiefly mammals and no dinosaurs. Mr. Gardner proposes to restudy this section. 28o STANTON assuming that all Ceratopsia everywhere must be above such an uncon- formity, and especially has this ceased to be valid reasoning since it has been proved that the Judith River Ceratopsia are in a formation intercalated between marine formations of the conformable Upper Cretaceous series. Physiographic Conditions Near the Close of the Cretaceous. Before discussing the paleontologic evidence of the age of the " Ceratops beds" it is desirable to consider briefly the relations of land and sea and the general conditions of sedimentation in the great Interior Region in late Cretaceous and early Tertiary time. It is well known that during the Benton epoch the sea covered practically all parts of that region in which Cretaceous rocks are now found. Only locally along the western margin were there estuaries and marshes in which coal was formed and from which purely marine life was excluded. Soon after the Benton, however, large areas west of the Front Range in Colorado and Wyoming and west of the loSth meridian in Montana previously covered by the sea began to emerge, either by uplift or by filling of the basins with sediment, and as they came up to sea level or a few feet above it land and marsh plants became established and all the con- ditions became favorable for the formation of coal beds. Land ani- mals also came in and the streams and fresh-water lagoons received their appropriate population from adjacent areas while the bays and estuaries were inhabited by brackish- water forms. Topographic con- ditions must have been such that sedimentation was practically con- tinuous from the marine deposits to the land and fresh-water sedi- ments. Such deposits are represented by the Mesaverde formation on the south and by the Eagle and Judith River formations on the north. The neighboring land masses must have formed large areas and have had considerable elevation in order to furnish the immense thickness of Upper Cretaceous sediments known in this region. The uplift was not uniform nor continuous. There were oscilla- tions so that occasionally brackish-water or marine deposits were brought above those of land and fresh-water origin, and it is probable that these oscillations were not always synchronous througiiout the region. Even in areas where marine waters did not come in for a long "CERATOPS beds" OF WYOMING AND MONTANA 28 1 time there must have been local subsidence about equalling the rate of deposition, because these thick formations seem throughout to have been formed near sea level. Locally, as in part of Bighorn Basin, this non-marine sedimentation may have been almost continuous until the end of the Cretaceous, but over most of the area there was a more important subsidence which brought the marine sediments represented by the Lewis and Bearpaw shales over the coal-bearing formations above-mentioned. Again the uplift was resumed and there was another transition from marine to land conditions. In some areas this was gradual with alternations of brackish- and fresh-water beds through a considerable thickness, as in southern Wyoming. There is a similar transition with less fresh-water sedimentation in the Laramie of the Denver Basin, where marine conditions had prevailed continuously since Benton time, and there is such a transition beneath the " Ceratops beds" of Converse County. In other areas the change is more abrupt and the marine rocks are followed directly by land and fresh-water deposits. The idea seems to be still prevalent that this latest Cretaceous sea was uplifted as a whole, its direct intercommunication with the ocean cut off, and its waters then gradually freshened so that its strictly marine animals were killed while those that could endure brackish waters survived. It has even been suggested that the change was so gradual that the brackish-water fauna may have adjusted itself to the changing conditions and continued to live after the waters became entirely fresh. Such a history seems to be implied in the statement of F. V. Hay den quoted on p. 250, and similar statements may be found in the writings of C. A. White. There is no evidence, however, in favor of this view. As far up in the series as brackish- water fossils are found they occur in usually thin beds intercalated amongst the fresh-water strata, showing that the two elements of the fauna had separate habitats. There is no more admixture of the two kinds of forms than is to be expected where slight oscillations of level alternately bring them over the same area and where currents may easily carry the fresh-water shells into brackish or even marine waters. "The intimate association of brackish- and fresh- water forms" may be found in printed lists and possibly in museum collec- tions but it does not exist in the rocks, and this is as true of the brack- ish-water shells of the highest horizon at which they occur in the Inte- 282 STANTON rior Region as it is of those in the Mesaverde and Judith River. These mollusks evidently lived in tidal waters connected somewhere with the open ocean. With all the oscillations the total result of the late Cretaceous move- ments was uplift^^ and this finally brought the whole Rocky Mountain and Great Plains regions permanently above sea level before the beginning of the Tertiary, for the nearest marine Eocene rocks now known are in eastern Texas, the lower Mississippi Valley, and on the Pacific Coast. Some areas in the Rocky Mountain region must have been subject to erosion for long periods before the close of the Cre- taceous. The hypothesis that erosion did not begin until the close of the Laramie in the Denver region, for example, and that time must be allowed between the Laramie and the Arapahoe for erosion to cut down through the 15,000 or 20,000 feet of sediments to the granite in that region is to my mind incredible and unnecessary. Age of the "Ceratops Beds." Evidence from the vertebrale fauna. — The vertebrate fauna of these beds is large and greatly varied, including mammals, dinosaurs, amphibians, rhynchocephalians, turtles, crocodiles, and fishes. As a whole and individually it is very closely related to the Judith River fauna, which is known to be Cretaceous and to lie beneath a thousand feet or more of marine Cretaceous beds. On this relationship with the Judith River fauna Brown^- says: The vertebrates are clearly of Mesozoic affinity. The dinosaurs here represented in the post-Laramie are the culmination of a practically uninterrupted line of highly organized vertebrates that have persisted with little change since the Judith River period, some like Claosaiirus, extending as far back as the Niobrara, and their relation to the earlier Jurassic forms is well established. *' I owe to Mr. Bailey Willis the suggestion that the facts would be better explained by an ebb tide of continental extent due to deepening of the ocean basins rather than by continental uplift. The fact of importance to my argument is the change in relative positions of land and sea, and whether this is due to continental uplift or to oceanic depression the resultant retreat of the sea from the interior of the continent is the same. "Bull. Am. Mas. Nat. Hist., Vol. XXIII, 1907, p. 845. " Williston, S. W. : The Laramie Cretaceous of Wyoming, Science, N. S., Vol. XVI, 1902, pp. 952-953. "CERATOPS beds" OF WYOMING AND MONTANA 283 Williston" states that this fauna has in some respects a startling resemblance to the Judith River fauna and argues that they were really contemporaneous. Perhaps the best general statement of the close relationship between these two faunas is that of Hatcher when he was bringing forward all the arguments he could find for their distinctness. The following quotations'* indicate his views, when it is remembered that he uses " Laramie" for the beds in Converse County. "When considered in its entirety, the vertebrate fauna of these (Judith River) beds is remarkably similar to, though distinctly more primitive than, that of the Laramie. Almost or quite all of the Lar- amie types of vertebrates are present, though, as a rule, they are repre- sented by smaller and more primitive forms. The similarity between this fauna and that of the Laramie contrasts strongly with the great dissimilarity between the vertebrates of the Judith River and those of the Atlantosaurus beds, the next older fresh-water horizon in this region, containing also a rich and varied fauna, but quite distinct from that of the beds in question. " Briefly, the Judith fauna, it is clear, is descended from the Juras- sic and is the direct ancestor of the Laramie. Its relations with the former are not close, and several groups are absent in the one which are present in the other. Its relations with the Laramie are much closer, as should be expected, considering the stratigraphic position. With one or two possible exceptions all the families represented in either of these two later deposits are present also in the other. Although several genera and species now appear to be common to both these formations, it is probable that when more perfect material is available they will be found, in most instances, to be quite distinct, though some pertaining to more persistent types may prove to be identical." Only two mammals have been reported from the Judith River for- mation and none from any other Upper Cretaceous horizon below the "Ceratops beds," and both of these according to Hatcher are closely related to forms in the "Laramie." According to Gidley there are two or three mammal genera in the " Cera tops beds" that range up into the Torrejon fauna of the Fort Union at Bear Butte, Montana, and one of these is represented by two species that may possibly be identical in both horizons. The turtles, rhynchocephalians, and crocodiles of the " Ceratops beds" may also be pretty closely related to those of the Fort Union, Puerco, and Torrejon, but the faunas are generally very distinct. Brown states''^ that the difference between the Puerco mam- " Bull. U. S. Geol. Survey, No. 257, 1905, pp. 101-103. " Op. cit., p. 840. 284 STANTON malian fauna and that of the "Laramie" is greater than that between the Puerco and Wasatch mammals. The Puerco formation of New Mexico lies between the Laramie and the Wasatch and contains two faunal zones. To the upper of these the vertebrate paleontologists have applied the nameTorrejon, restrict- ing Puerco to the lower one. It has been shown that in Montana the Fort Union formation includes mammal-bearing beds with a fauna more closely related to the Torrejon fauna than to the restricted Puerco but this probably does not mean that only the upper zone is strati- graphically represented. It is fair therefore to say that the Puerco as a whole is in Montana stratigraphically above the " Ceratops beds." The Puerco is now generally treated as basal Eocene, though Cope called it Mesozoic and placed it with the Laramie in a separate sys- tem beneath the Tertiary, and if it is basal Eocene then the Ceratops beds should be Cretaceous on that showing alone. Osborn^* says of the restricted Puerco that its mammalian fauna is " wholly of Mesozoic origin, and mostly destined to disappear; not a single representative or ancestor of any existing orders of Tertiary mammals is certainly known" and that it has no representative in Europe. Concerning the Torrejon mammals he says that out of 40 species only one is of "modernized Tertiary stock," all the others being Mesozoic, i.e., derived from and related to the Mesozoic. The Torrejon is compared with the Thanetien or Cernaysien stage of France. Apparently it would do less violence to the vertebrate e\i- dence to put the Puerco and Torrejon horizons in the Cretaceous than to put the "Ceratops beds" in the Tertiary. In France the Montien stage now referred to the uppermost Cre- taceous contains dinosaurs closely related to those of the Tricera- tops fauna, according to De Lapparent.^^ The vertebrate evidence for Cretaceous age does not rest on a single dinosaur or a few members of that group associated with a Tertiary fauna. If this were the case it would be fair to argue that they were straggling descendants of an earlier fauna that had lived beyond their normal time. The whole fauna is Cretaceous in its affinities and the fair question is not could the dinosaurs have lived on into the Tertiary, *" Osbom, H. F. : Cenozoic mammal horizons of Western North America' Bull. U. S. Geol. Survey, No. 361, 1909, pp. 34, 35. '•' Traits de Geologic, 5™^ ed., 1906, p. 1472. "CERATOPS beds" OF WYOMING AND MONTANA 285 but did they do so ? The fact that ammonites, Inocerami, rudistae, and many other groups of invertebrates have no descendants in the Tertiary does not disqualify them as characteristic fossils of the Meso- zoic, neither does the fact that the Cretaceous dinosaurs left no Tertiary descendants disqualify them as characteristic Mesozoic fossils. Evidence from the invertebrate fauna. — The two elements of the inver- tebrate fauna, one inhabiting the brackish waters and the other the fresh waters, have already been discussed and the fact has been empha- sized that their local habitats were really distinct, though they might be near each other and might alternately occupy the same territory many times. It is obvious then that an attempt to correlate one locality where there were only fresh-water deposits with another locality where the beds were chiefly of brackish-water origin by count- ing identical species and figuringpercentages would lead to erroneous results. A comparison of the Laramie invertebrates of the Denver Basin with the invertebrates of the " Ceratops beds" of Hell Creek and Converse County is a case of this kind and is of no value whatever. A similar result would be obtained by comparing the fauna of the Poto- mac at Washington with that at the mouth of the same stream. At Washington, although within tidal waters, there is a varied fresh- water fauna, which includes many of the genera that lived in Creta- ceous time, while in the lower course of the river and Chesapeake Bay there are beds of oysters and the few other brackish-water forms usually associated with them. The invertebrates of the "Ceratops beds" and indeed practically all of the non-marine forms of the Cretaceous like the plants of the Upper Cretaceous belong to living genera and, taking the world over, it is probable that almost every specific type may be found represented by similar living species. It must be admitted that in themselves, without any reference to stratigraphic occurrence or local geologic history, these fossils could not be depended upon for the discrimina- tion of horizons within the Cretaceous nor for distinguishing between Cretaceous and Tertiary. But when the investigation is confined to a single region and when the geographic and stratigraphic range of non-marine species has been determined their evidence is useful and important. When the brackish-water fauna of a locality like Point of Rocks, Wyoming, is studied and is proved to be of Cretaceous age by means of overlying marine faunas, and when a very similar assemblage 286 STANTON with several identical forms is found at a much higher horizon at Black Buttes,afew miles away, it is a fair inference that the Black Buttes fauna is directly descended from that at Point of Rocks and that both are of Cretaceous age. When the whole region is studied and it is learned that no marine Tertiary fauna ever entered it and that the Tertiary sea did not approach closely enough to the region to have been connected with these brackish waters, and when it is further learned that no such assemblage of non-marine forms is anv- where found intercalated in or closely associated with marine Tertiary rocks the truth of the inference that the brackish-water fossils of Black Buttes are Cretaceous is practically demonstrated and it may be confidently asserted that in this area all higher rocks in which such fossils occur are also Cretaceous. Of course isolated occurrences of Corhula and perhaps a few other genera that are known to range into purely fresh waters far from the ocean must be treated with caution. It is for these reasons that I have given much importance to the oyster bed reported by Leonard in southwest North Dakota above the Tri- ceratops horizon and above an unconformity. It is in itself proof that the rocks are Cretaceous and that the unconformity has no significance in the matter of separating Cretaceous from Tertiary. The brackish- water fauna of the Upper Cretaceous has the same general character and many specific identities throughout its great vertical and geogra- phic range. The fresh-water mollusca include a few species that actually pass up into the overlying formation, but the majority of the species are confined to this horizon or occur in either identical or closely related forms in the Mesaverde, Judith River, and other Cretaceous forma- tions. This element of the fauna is best developed in the Hell Creek and Converse County areas, where they are not associated with the brackish-water beds, but the distribution of some of the species is wide-spread in association with the dinosaur fauna, and a large pro- portion of them, including some of the more striking and characteris- tic forms, occur at Black Buttes in beds that contain intercalated brackish-water strata. These facts together with the absence of types similar to most of the forms from the Eocene and other Tertiary rocks of the region justify the reference of the frcsh-watcr fauna to the Cretaceous. Evidence from the flora. — Doctor Knowlton's statistics show a close "CERATOPS beds" OF WYOMING AND MONTANA 287 relationship between the tlora of the "lower Fort Union" and the "upper Fort Union;" if, however, comparison were made between the "Ceratops beds" that have actually yielded dinosaurs and the typical, or "upper" Fort Union, the figures would possibly be different, because ir many sections such as those of the Bighorn Basin, Red Lodge, Sheridan, Fish Creek, and probably along the Yellowstone, he has included much more than the "Ceratops beds" in his lower Fort Union. The value of the comparison with the Laramie flora is impaired also by the fact that it is restricted to the 80 species of the Denver Basin Laramie which is probably only a fraction of the whole flora. Even with that restriction 10 species, or 12^ per cent of the Laramie fiora, occur in the "lower Fort Union" and only about the same percentage of " upper Fort Union" species occur in it, though they are 61 in number. It is of course true that a considerable num- ber of Fort Union plants are found in the "Ceratops beds," but that does not necessarily prove that the latter are Eocene, nor that they belong to the Fort Union formation. The difference between the Laramie and Fort Union floras was first emphasized by Dr. J. S. Newberry, who studied and described^^ the early collections of Fort Union plants which he then considered Mio- cene. The Fort Union known to him, as indicated by the localities, is all "upper Fort Union." His final views concerning the Fort Union and Laramie floras are shown by the following extracts from I wo of his latest papers, which also clearly show that the "Laramie" as known to Newberry included Colorado, Mesaverde, Laramie, Denver, and possibly later beds :" Between the foot hills of the Rocky mountains and the summit of the Wasatch the older Cretaceous rocks are covered by sandstones and shales which contain beds of coal so numerous and important as to give to the formation the name, sometimes applied to it, of the Lignitic Forma- tion. This, the true Laramie group, is distinct from the so-called Laramie of the upper Missouri, named by Hayden the Fort Union group, and which, in my judgment, should be considered the basal member of the Tertiary. Large collections of fossil plants made by Dr. Hayden and others in the country bordering the upper Missouri ** Annals Lyceum Nat. Hist. N. Y., Vol. 9, 1868, pp. 1-76. " The Coals of Colorado, School of Mines Quarterly, Vol. IX, 1888, p. 329. The Laramie group, its geological relations, its economic importance and its fauna and flora, N. Y. Acad. Sci., Vol. IX, 1889, pp. 27, 31, 32. Proc. Wash. Acad. Sci., August, 1909. 288 STANTON have passed through my hands, and have been described by me in a memoir to be pubhshed by the U. S. Geological Survey. This flora has also been carefully studied by Prof. L. F. Ward, who has published an important paper upon it which forms part of the Sixth Annual Report of the Director of the U. S. Geological Survey. It is entitled "A Synopsis of the Flora of the Laramie Group," but it really contains few of the Lower or true Laramie plants, and is chiefly descriptive of the Upper Laramie or Fort Union flora, of which it affords the most important review yet published. In this flora are a number of living species, such as Corylus americana, C. rosirata, Onoclea sensihilis, etc., and many species found in the Tertiary rocks of northern North America, Greenland, and Europe, in strata called Micoene by Heer, but shown by J. Starkie Gardner to be Eocene. The flora of the Lower or true Laramie has been made known to me by collections from Evanston, Black Butte, Bitter Creek, Fisher's Peak, Walsenburg, Canon City, Crested Butte, etc. These show that there is almost nothing in common between the Fort Union and Lara- mie floras, and that the two divisions of the so-called Laramie group, judging from their fossil plants, which are very numerous, must be regarded as distinct formations. The older flora contains no living species, but has many things in common with the Upper Cretaceous coal-bearing rocks of Vancouver Island and Washington Territory, is associated with marine and brackish water shells among which Inoceramus is conspicuous, and I have, therefore, called it Cretaceous and the upper member of that system. The following is from the second paper cited: The term Laramie Group was first used by Mr. Clarence King, who applied it to a series of sandstones and shales containing beds of coal exposed at Point of Rocks, Black Butte, and other places along the line of the Union Pacific Railroad, and having great development in Colorado and Wyoming. Mr. King considered this the uppermost member of the Cretaceous system, and excluded from it the Fort Union beds of Hayden, which, on the evidence I furnished him, he agreed with me in considering Tertiary. I called the Fort Union group Miocene because I identified it with the plant-bearing beds of Mackenzie's River, Disco Island, Greenland, etc., of which the flora had been studied by Prof. Oswald Heer and was by him called Miocene. This flora, to which I shall again refer, has since been shown by Mr. J. Starkie Gardner to be Eocene. The Fort Union flora has many species in common with the Eocene beds of the Island of Mull, Bournemouth, etc., and holds undoubtedly the same geological posi- tion. Dr. Hayden accepted the term Laramie Group, but made it include his Fort Union beds. "CERATOPS beds" OF WYOMING AND MONTANA 289 The Laramie group is a formation that is essentially confined to the Rocky Mountain region; it forms a marginal belt on the east side of the mountains, extending from Central Mexico far into the British possessions. On the west side of the Rocky Mountains, it stretches over to the Wasatch, but has not been recognized at any point west of the summit of that range. Everywhere it contains coal, frequently in large quantity and of excellent quality. These coals are opened along the eastern side of the Rocky Mountains at Erie, Marshall's, Florence, Walsenburg, Trinidad, etc.; and from these sources the rapidly growing towns of the prairie region are receiving most of their fuel. On the west side of the Rocky Mountains, in Colorado, the Laramie coals are more important than elsewhere; the aggregate thickness in some places of several beds is fifty feet or more; the coal varies in composition from hard anthracite to open-burning bitumin- ous, and much of it is of excellent purity, the amount of ash ranging from 2 to 5 per cent, the sulphur often less than half of one per cent. It is from this formation that coal is taken at various points along the Union Pacific Railroad at Coalville, Pleasant Valley, and Castle Valley in Utah Most of the so-called Denver Tertiary beds really belong to the Laramie group, at least those beds which contain the remains of Ceratops and the stratified ash-beds of South Table Mountain. These represent the upper part of the formation, and fully one-half of the fossil plants contained in them are also found in the Lower Laramie at Golden, Florence, Trinidad, etc. In Europe, the formations which indicate by their fossils the nearest approach to the Laramie are the Paleocene plant-bearing beds of Sezanne, Gehnden, and Alum Bay. It is not known that any species found in these localities is represented in the Laramie flora; and in the absence of all connecting links, we should be unwarranted in calling the Laramie Paleocene from its fossil plants, while its moUusks and vertebrates forbid this. The floras of Sezanne and.Gelinden are cited by Knowlton as re- lated to the Fort Union flora and this relationship is one of the argu- ments for referring the latter to the Eocene. Lesquereux*" compares the flora of Point of Rocks, now known to be Cretaceous, with the same European Eocene floras, as well as with that of Black Buttes. He says: Of the thirty species enumerated in the table, two appear identical with, and one is related to, Canadian species, recognized as Tertiary, "Tertiary Flora, U. S. Geol. Survey Terr., Vol. VII, 1878, pp. 342-344. 347-351- 290 STANTON as seen below from quotations of Prof. G. M. Dawson's report. Six are identical with, and seven are analogous to, those of the Lower European Miocene; two are identical with, and one allied to, Arctic Miocene" species. Seven have a close relation to plants of the Lower European Eocene, Sezanne and Gelinden, two localities composing a subdivision separated at the base of the Tertiary, under the name of Paleocene. Three are identified and one analogous in the flora of Golden; nine identical and one analogous in that of Black Buttes; and four have analogy with Cretaceous forms The groups of plants at Point of Rocks has, besides the Eocene representatives, six species identified with, and as many related to, those of the Miocene of Europe. Therefore, we see here, what has been remarked in other localities of the Lignitic, a compound or admix- ture of old and young Tertiary types, in comparison at least with the fossil floras of Europe, and thus a general character which does not distinctly relate our first group to any peculiar stage of the European Tertiary. We have the Paleocene by relation to species of Gelinden and Suzanne; the Upper Eocene, especially the Ligurian, or Oligocene, by the Palms; and the Miocene by a number of common and generally distributed forms, which, like Sequoia hrevifolla, Sequoia langsdorfii, Populus mutabilis, Ficus tilicefolia, Rhamnus rectinervis, Juglans rugosa, etc., are persistent types of wide distribution, indicating merely the Tertiary age for the Lignitic flora. For this reason, I shall continue to carefully record its points of affinity with the divers groups of the geological floras of Europe; but at the same time, denying as yet sufficient evidence for its identification to any of them, I persist in considering it simply as the Lower Eocene flora of this continent. Mr. J. Starkie Gardner, who is cited by Newberry as authority for the reference to the Eocene of certain Arctic American floras that are related to the Fort Union flora, had such peculiar views concerning the characteristics of Cretaceous floras that his opinions on correla- tions of formations outside of Great Britain have little weight. His study of British floras led him to conclude that there are no Cretace- ous dicotyledons.^- He therefore regarded the floras of the Dakota sandstone, of Vancouver Island, and of many European localities usually called Cretaceous and " Knowlton's review of the Montana flora, including the Point of Rocks species (Bull. U. S. Geol. Survey, No. 163, iqoo), does not recognize close relationship with Tertiary floras and discredits most of Lesquereux's identi- fications in it of European Eocene and Miocene species. "Gardner, J. Starkie.: On the age of the Laramie formation as indicated by its vegetable remains, Am. Naturalist, Vol. XIV, 1880, pp. 565-560. CERATOPS beds' OF WYOMING AND MONTANA 291 characterized by an abundance of dicotyledons, as belonging to a vast intermediate period which he called Cretaceo-Eocene. Speaking of the evidence for such an intermediate period he says: "Taking the floras of Cretaceous age in England, whose horizon is absolutely known, we see that they point to an even greater interval (between Cre- taceous and Tertiary) than the fauna. Of course I leave out of the question the so-called Upper Cretaceous floras of Europe whose age, not based upon stratigraphic evidence, is even more a matter of doubt than those of America. Later Gardner'^ again questioned the Cretaceous age of the Dakota flora. The following quotation has a bearing on the present discus- sion: The Cretaceous series of America contains at its very base a flora composed of angiosperms so perfectly differentiated that they are apparently referable to existing genera. One of the oldest floras in Europe containing angiosperms is that of Aix-la-Chapelle, and even this we have seen is relatively modern; but these are not referable in at all an equal degree to existing genera, and even the coniferae are embar- rassing on account of their highly transitional character. The oldest Cretaceous flora" of America so far from possessing any Cretaceous characters, agrees in a remarkable manner with that of the English Lower Eocene, while the Laramie, or supposed Cretaceo-Eocene, flora has very much in common with that of our Middle Eocene, and marks a similarly sudden rise in temperature. Gardner's views on the relationships of the Dakota and Laramie floras are presumably not now held by any paleobotanist. His statements are quoted here merely for the purpose of emphasizing the fact that in Europe there is no well-established standard series of Upper Cretaceous floras with which comparisons can be made and that too little is known about the geologic relations of the Arctic American floras to make them available for that purpose. It is prob- able that the most complete succession of Cretaceous and Tertiary floras in the world is preserved in the rocks of the United States and Canada, and when these are fully studied in all their biologic and stratigraphic relations they will form the standard for comparison and correlation. ** On the relative ages of the American and English Cretaceous and Eocene series, Abstracts in Nature, Vol. XXX, 1884, pp. 538-529. Rept. Brit. Assoc. Adv. Sci., Montreal, 1884, pp. 739-741. °* The Dakota flora was the oldest known at that time. 292 STANTON It is obvious from the testimony of the paleobotanists, and this testimony could be indefinitely increased by similar examples, that the modern flora in its essential features was well developed in Creta- ceous time at least as early as the Dakota epoch (or approximately the Cenomanian of European geologists). The changes since that time have been mostly due to minor differentiations of the types then intro- duced. The floras therefore do not furnish competent evidence for discriminating between Cretaceous and Tertiary. These later floras are useful in local stratigraphy and in provincial or regional correlation in much the same way that the non-marine faunas are, though they are usually more important than the non-marine faunas because they are more numerous and more highly differentiated. They cannot be used in determining the boundary between Cretaceous and Eocene until the stratigraphic range of the species is thoroughly known and the evidence of the plants is checked by means of other criteria. The differences between late Cretaceous and early Tertiary floras are chiefly specific characters and modifications in the grouping of the types, both of which must have varied from place to place as well as from time to time. The method, often adopted by paleobotanists, of determining the relative stratigraphic positions of local fossil floras by ignoring all the undescribed and unidentified species, and reckoning percent- ages of identified species that are recorded in various horizons, is based on the unwarranted assumption that there were no variations in topography, soil, or other conditions affecting plants, and that the distribution and preservation of the species was uniform through- out their range. That this method should lead to errors in correla- tion is inevitable. Conclusions. In the Interior Region of North America the formations between the uppermost marine Cretaceous and the Wasatch together constitute a real transition from the Cretaceous to the Tertiary. Notwithstanding the fact that there are several local unconformities at various horizons and perhaps some of more general distribution there is no conclusive evidence that any one of these represents a very long period of erosion not represented by sediments elsewhere in the region. CERATOPS BEDS OF WYOMING AND MONTANA 293 The Fort Union formation, properly restricted, is of early Eocene age, the determination resting chiefly on its stratigraphic position and its primitive mammalian fauna which is related to the earliest I^ocene fauna of Europe. The very modern character of the flora tends to confirm this correlation. The "Ceratops beds" are of Cretaceous age as decided by strati- graphic relations, by the pronounced Mesozoic character of the verte- brate fauna with absence of all Tertiary types, and by the close rela- tions of its invertebrate fauna with the Cretaceous. The relations of the flora with Eocene floras is believed to be less important than this faunal and stratigraphic evidence. Taken in their whole areal extent they probably include equivalents of the Laramie, Arapahoe, and Denver formations of the Denver Basin. INDEX. Note. — New names in black-face type, synonyms in Italics. abrupta, Dicerca 140 acerba, Dicerca 15S acomana, Buprestis loi acutangulus, Gyascutus 63 adducta, Buprestis 103 adjecta, Buprestis 118 adonea, Buprestis 98 adulans, Buprestis 123 aemula, Buprestis 121 aeneoviridis, Gyascutus obesus 66 aerata, Gyascutus amplus 64 aestiva, Dicerca 146 affinis, Buprestis 123 Age and stratigraphic relation of the Ceratops Beds of Wyoming and Montana 239 altemans, Buprestis 93 American Buprestidas 47 americana, Dicerca 163 amplicata, Dicerca 153 amplus, Gyascutus 64 Anataxis 170 gentilis 172 angusta, Buprestis 108 Ancylocheira 87 angulicollis, Chalcophora 80 Anoplis 87 angusticauda, Dicerca 148 angustus, Hippomelas (Prasinalia) 59 apricans, Buprestis 128 arizonica, Spinthoptera 74 austera, Dicerca lugubris 157 aurichalcea, Dicerca 145 biangulata, Dicerca 140 bifoveata, Dicerca 159 bisinuata, Texania 87 bistrinotata, Buprestis 108 boulderensis, Buprestis torva 96 brevicollis, Chaleophora lacustris 79 brevipes, Hippomelas 56 brevis, Buprestis (Cypriacis) 118 Buprestidae, American 47 Genera defined 48-53 Buprestis 51; 87 acomana loi adjecta 118 adducta 103 adonea 98 adulans 123 aemula 121 affinis 123 alternans 93 conicicauda 93 angusta 108 apricans 128 bistrinotata 108 brevis 118 caliginosa 102 callida 109 canadensis 124 confiuenta 104 tessellata 104 connexa 1 i i consularis 96 contorta 95 crenata no leviceps in seditiosa no cribripennis 127 decora 127 depressa in diruptans 94 elongata 106 fabulosa 120 fasciata 107 fortunata 107 fastidiosa 109 mediocris 109 flavopicta 96 fulgens 107 fusca 1 01 sublivida 102 gibbsi 108 graminea 114 gravidula 95 boulderensis 96 torva 96 histrio 98 inconstans 9 1 scripta 91 impedita 124 intricata 119 laeviventris 93 pugetana 94 langi 112 lateralis 119 lanta 120 leporina 92 lineata 89 295 296 INDEX Buprestis lyrata 103 maculativentris 100 maculipennis 90 deficiens 91 fusiformis 91 morosa loi nigricans 102 nupta 122 nuttalli 92 obscura 125 ornata 112 incolumis 112 obliqua 113 oregonus 113 radians 121 rubronotans 97 sexplagiata 106 saturata 114 striata 125 subornata 98 punctiventris 99 sulcicollis 119 tacomae 121 prospera 123 ultramarina 117 venusta 122 villosa 114 violescens 99 viridimicans 112 Ccelata, Stictocera 69 Calcite crystals with new forms i from Andreasberg i from California 6 from England 2 from New Mexico 8 from Rhode Island 11 from Texas 14 califomica, Nanularia 174 callida, Buprestis 109 caliginosa, Buprestis 102 callosa, Dicerca 150 campestris, Texania 85 canadensis, Buprestis 124 carolinensis, Gyascutus 66 Casey, Thos. L. 47 caudata, Dicerca 142 Ceratops Beds of Wyoming and Montana 239 Chalcophora 51; 77 anguilicoilis 80 montana 80 fortis 82 georgiana 81 hondurasica 79 iridescens 82 lacustris 79 brevicoUis 79 laurentica 82 liberta 83 oregonensis 80 parviceps 83 prominens 81 virginiensis 78 obliterata 79 chrysea, Dicerca 159 Cinyra 174 gracilipes 176 macilenta 176 compactus, Gyascutus 66 confluenta, Buprestis 104 conicicauda, Buprestis alternans 93 connexa, Buprestis in consularis, Buprestis 96 contorta, Buprestis 95 convexa, Spinthoptera 73 cuneatus, Hippomelas (Prasinalia) 59 cupreofusca, Nanularia 174 cupreola, Dicerca 141 cupreopunctata, Spinthoptera 75 cupripes, Pcecilonota 169 crassicollis, Dicerca 155 crenata, Buprestis no cribriceps, Gyascutus 62 cribripennis, Buprestis 127 Cross, Whitman, 27 Crystals with new forms, Some Cal- cite I cyanipes, Pcecilonota 168 cylindricus, Hippomelas 58 cylindrinus, Gyascutus 67 Cypriacis 116 debilis, Gyascutus 67 Pcecilonota 170 decora, Buprestis 127 depressa, Buprestis in Dicerca 52; 128 abrupta 140 acerba 158 aestiva 146 americana 163 angusticauda 148 aurichalcea 145 biangulata 140 bifoveata 159 mystica 159 callosa 150 caudata 142 chrysea 159 crassicollis 155 hesperica 155 cupreola 141 distinguenda 157 divaricata 144 incisa 144 dumolini 157 erosa 137 filiola 142 floridae 134 gaudens 136 gracilis 133 INDEX 297 Dicerca liorni 153 amplicata 153 indurata 131 inflatula 140 innocua 133 lepida 138 levettei 135 longicauda 141 lugubris 157 austera 157 lurida 136 montana 160 mono 156 mutica 165 nigra 146 obscura 131 parumpunctata 145 pectorosa 165 pisciformis 141 porcatula 133 prolongata 149 longipennis 150 pugionata 138 punctulata 154 pinorum 154 rigida 151 rustica 146 sagax 134 scobina 164 seriata 136 severa 149 subaequalis 143 subargentea 148 subcuprea 147 pertinax 147 soror 134 regularis 134 stolida 161 sulcatula 147 tenebrica 149 tenebrosa 155 tetrica 151 truncata 132 tuberculata 158 vancouveri 148 wickhami 160 deficiens, Buprestis maculipennis 91 diruptans, Buprestis 94 distinguenda, Dicerca 157 divaricata, Dicerca 144 dumolini, Dicerca 157 egregius, Gyascutus 64 elongata, Buprestis 106 erecta, PcEcilonota 167 erosa, Dicerca 136 fabulosa, Buprestis 120 Fall, H. C. 47 fasciata, Buprestis 107 fastidiosa, Buprestis 109 ferrea Poccilonota 167 fldelis, Gyascutus 65 filiola, Dicerca cupreola 142 flavopicta, Buprestis 96 floridae, Dicerca 134 fortis, Chalcophora 82 fortunata, Buprestis fasciata 107 Fort Union Formation 181 fulgens, Buprestis 107 fulleri, Texania 86 fusca, Buprestis loi fusiformis, Buprestis maculipennis 9 1 gaudens, Dicerca 136 gentilis, Anataxis 172 georgiana, Chalcophora 81 gibbsi, Buprestis 108 gracilipes, Cinyra 176 gracilis, Dicerca 133 graminea, Buprestis 114 gravidula, Buprestis 95 Greene, Edward L. 17 grossus, Hippomelas 57 Gyascutus 50; 60 acutangulus 63 amplus 64 asrata 64 carolinensis 66 compactus 66 solidus 66 cribriceps 62 cylindrinus 67 debilis 67 egregius 64 fidilis 65 juniperinus 68 obesus 66 seneoviridis 66 obliteratus 65 pimalicus 63 pistorius 63 planicosta 61 snowi 62 tenuis 67 Hell Creek Beds 179 hesperica, Dicerca crassicollis 155 Hippomelas 50; 55 angustus 59 brevipes 56 cuneatus 59 cylindricus 58 grossus 57 metallicus 60 mexicanus 58 planicauda 58 robustus 57 serrulatus 57 sphenicus 57 histrio, Buprestis 98 hondurasica, Chalcophora 79 Horn, Dr. G. H. 47 homi, Dicerca 153 INDEX incisa, Dicerca 144 incolumis, Buprestis omata 112 inconstans, Buprestis 91 indurata, Dicerca 131 inflatula, Dicerca 140 innocua, Dicerca 133 insularis, Spinthoptera 76 iridescens, Chalcophora 82 intricata, Buprestis 119 impedita, Buprestis 124 juniperinus, Gyascutus 68 Knowlton, F. H. 179 Knaus, Warren 49 lacustris, Chalcophora 79 laeviventris, Buprestis 93 langeri, Texania 87 langi, Buprestis 112 Laramie formation 27 lateralis, Buprestis 119 laticornis, Stictocera 70 lauta, Buprestis 120 laurentica, Chalcophora 82 Leng, Chas. W. 49 lepida, Dicerca 138 leporina, Buprestis 92 levettei, Dicerca 135 leviceps, Buprestis in limula, Dicerca divaricata 144 liberta, Chalcophora 83 lineata, Buprestis 89 Linnajus as an evolutionist 17 Linnaius to Gmelin 25 longicauda, Dicerca 141 longipennis, Dicerca prolongata 150 lugubris, Dicerca 157 lurida, Dicerca 136 lyrata, Buprestis 103 macilenta, Cinyra 176 maculativentris, Buprestis 100 maculipennis, Buprestis 90 Manee, A. H., 49 mediocris, Buprestis fastidiosa 109 metallicus, Hippomelas(Prasinalia)6o montana, Chalcophora angulicollis 80 montana, Dicerca 160 morio, Dicerca 156 morosa, Buprestis loi mutica, Dicerca 165 mystica, Dicerca 159 Nanularia 172 californica 174 cupreofusca 174 New forms of Calcite Crystals i nigra, Dicerca 143 nigricans, Buprestis 102 nupta, Buprestis 122 nuttalli, Buprestis 92 obesus, Gyascutus 66 obliqua, Buprestis oregona 113 obliterata, Chalcophora virginiensis 79 obliteratus, Gyascutus 65 obscura, Buprestis 125 obscura, Dicerca 131 ocularis, Spinthoptera 73 oregona, Buprestis incolumis 113 oregonensis, Chalcophora 80 omata, Buprestis 112 Paleontology, etc., of the Hell Creek Bed, etc., 179 paganorum, Buprestis 100 parva, Spinthoptera 74 parviceps, Chalcophora S3 parumpunctata, Dicerca 145 parviceps, Poecilonota 169 patruelis, Buprestis 113 pectorosa, Dicerca 165 pertinax, Dicerca subcuprea 147 pimalicus, Gyascutus 63 pinorum, Dicerca 154 pisciformis, Dicerca 141 pistorius, Gyascutus 63 planicauda, Hippomelas 58 ?lanicosta, Gyascutus 61 oecilonota cupripes 169 cyanipes 168 debilis 170 erecta 167 ferrea 167 parviceps 169 thureura 170 pollens, Stictocera 70 porcatula, Dicerca 133 Prasinalia, new subgenus 59 prolongata, Dicerca 149 prominens, Chalcophora 81 prospera, Buprestis 123 pugetana, Buprestis laeviventris 94 pugionata, Dicerca 138 punctiventris, Buprestis subomata 99 punctulata, Dicerca 154 radians, Buprestis 121 reducta, Buprestis 92 regularis, Dicerca soror 135 rigida, Dicerca 151 robustus, Hippomelas grossus 57 rubronotans, Buprestis 97 rufipes, Buprestis 105 rustica, Dicerca 146 rusticorum, Buprestis 99 sagax, Dicerca 134 saturata, Buprestis 114 Schaller, Waldemar T. i scobina, Dicerca 164 scripta, Buprestis inconstans 91 seditiosa, Buprestis crenata no seriata, Dicerca 136 INDEX 299 serriger, Texania 86 serrulatus, Hippomelas 57 severa, Dicerca 149 sexplagiata, Buprestis 106 Shoshone group 27 snowi, Gyascutus 62 solidus, Gyascutus compactus 66 sorer, Dicerca 134 Spectralia 175 sphenicus, Hippomelas 57 Spinthoptera 50; 71 arizonica 74 convexa 73 cupreopunctata 75 insularis76 ocularis 73 parva 74 valens 73 webbi 7 5 woodhousei 74 Stanton Timothy W. 239 Stereosa, 126 Stictocera 50; 69 cselata 69 laticornis 70 pollens 70 stolida, Dicerca 161 Stratigraphic relations of the Cera- tops Beds of Wyoming and Montana 179 striata, Buprestis 125 Studies of American Buprestidce 47 subaequalis, Dicerca 143 subargentea, Dicerca 148 subcuprea, Dicerca 147 sublivida, Buprestis fusca 102 subomata, Buprestis 98 sulcatula, Dicerca 147 sulcicoUis, Buprestis 119 tacomae, Buprestis 121 tenebrica, Dicerca 149 tenebrosa, Dicerca 155 tenuis, Gyascutus 67 tessellata, Buprestis confluenta 104 tetrica, Dicerca 151 Texania 51; 84 bisinuata 87 campestris 85 fulleri 86 langeri 8 7 serriger 86 thureura, Poecilonota 170 torva, Buprestis 96 Trachykele 177 truncata, Dicerca 132 tuberculata, Dicerca 158 ultramarina, Buprestis 127 vancouveri, Dicerca 148 valens, Spinthoptera 73 venusta, Buprestis 122 villosa, Buprestis 114 violescens, Buprestis 99 virens, Buprestis 105 virginiensis, Chalcophora 78 viridimicans, Buprestis 112 webbi, Spinthoptera 75 Wickham, H. F. 49 wickhami, Dicerca 160 woodhousei, Spinthoptera 74 MBL WHOI Llbrarv Sonnls iiiililii iiii'iii' ilil'M 5 WHSE 00872 n I I i ^»i