Hoploscaphites gilberti, Landman & Kennedy & Cobban & Larson & Jorgensen, 2013

Landman, Neil H., Kennedy, W. James, Cobban, William A., Larson, Neal L. & Jorgensen, Steven D., 2013, A new species of Hoploscaphites (Ammonoidea: Ancyloceratina) from cold methane seeps in the Upper Cretaceous of the U. S. Western Interior, American Museum Novitates 2013 (3781), pp. 1-40 : 11-35

Hoploscaphites gilberti , n. sp.

Figures 4–13

Scaphites nodosus . Gilbert, 1896: pl. 63, fig. 3.

Hoploscaphites gilli . Cobban and Jeletzky, 1965: pl. 95, fig. 1A–D.

DIAGNOSIS: Adult shell compressed, with rounded to elongate outline in lateral view; strongly dimorphic; macroconchs with relatively short shaft and recurved hook leaving almost no gap between exposed phragmocone and hook; apertural angle averages 50°; umbilical seam straight in lateral view; microconchs with longer shaft and more recurved hook leaving larger gap between exposed phragmocone and hook; umbilical seam concave in lateral view; in both dimorphs, intercostal whorl section of shaft subquadrate with fairly flat, subparallel flanks and broadly rounded venter; ribs slightly flexuous with weak adoral projection on venter; umbilicolateral bullae or raised ribs on body chamber; ventrolateral tubercles on exposed phragmocone and body chamber, usually extending to aperture; suture simple with broad, asymmetrically bifid first lateral saddle and narrow, symmetrically to asymmetrically bifid first lateral lobe.

ETYMOLOGY: For Grove Karl Gilbert, who first illustrated this species in 1896 in his seminal study in which he first described the tepee buttes of the Pierre Shale of Colorado.

TYPES: Holotype is AMNH 64559 , a macroconch, from the Baculites scotti Zone, Pierre Shale, AMNH loc. 3386, Butte County, South Dakota . Paratypes are USNM 547296 , a macroconch, from the Baculites scotti Zone, Pierre Shale, USGS Mesozoic loc. D3464, Pueblo County, Colorado ; USNM 547293 , a macroconch, from the Baculites scotti Zone, USGS Mesozoic loc. D84, Pueblo County, Colorado ; USNM 28362 , a microconch, from the Pierre Shale , USGS loc. 1359, Pueblo County, Colorado , as illustrated by Gilbert (1896: pl. 63, fig. 3); AMNH 64573 , a

TABLE 2 Measurements of Hoploscaphites gilberti , n. sp., macroconchs See figure 3 for description of measurements. All measurements are in mm except for apertural angle, which is in degrees. * = Pierre Shale, Fall River County, South Dakota.

Specimen Locality Zone LMAX LMAX/ LMAX/ Apt UD UD/ WP/ WS/ WH/ VS/ Number HP HS <LMAX HP HS HH HS AMNH 64531 AMNH B. scotti – – – 73.0 – – – 0.86 1.07 0.58

3386

AMNH 64559 AMNH B. scotti 32.0 3.23 2.34 42.5 3.38 0.11 0.92 0.83 1.02 0.65 3386

AMNH 64561 AMNH B. scotti 35.2 3.09 2.11 – 2.02 0.06 0.87 0.78 0.96 – 3386

AMNH 64587 AMNH B. scotti 32.6 3.26 2.28 51.5 3.46 0.11 0.90 0.76 1.07 0.60 3386

AMNH 64601 AMNH B. scotti 40.5 3.21 2.45 47.5 3.95 0.10 0.87 0.82 1.01 – 3386

AMNH 64646 AMNH B. scotti 37.0 3.27 – – 3.58 0.10 0.92 – 0.97 – 3386

AMNH 64699 AMNH B. scotti 55.2 3.27 2.05 71.5 5.61 0.10 1.09 0.90 – 0.59 3386

AMNH 81100 AMNH B. scotti 41.8 3.40 2.34 45.0 3.79 0.09 1.02 0.80 1.06 0.64 3386

AMNH 81104 AMNH B. scotti 46.7 3.22 2.07 – 4.72 0.10 0.96 – – – 3386

AMNH 81117 AMNH D. nebras- 53.7 2.97 – 55.0 3.90 0.07 0.81 – 1.01 – 3340 cense

AMNH 81443 AMNH B. scotti 37.7 3.06 – – 3.61 0.10 0.89 – 1.04 – 3386

AMNH 81469 AMNH B. scotti 45.2 – – – – – – – – – 3386

AMNH 81473 AMNH B. scotti 37.2 3.35 2.42 – 5.08 0.14 0.90 – 1.05 – 3386

AMNH 81478 AMNH B. scotti 32.1 3.00 – – 3.40 0.10 0.92 – 1.11 – 3386

AMNH 83717 AMNH B. scotti 45.0 3.04 2.07 74.0 – – 0.89 0.74 1.03 0.53 3386

Sc 590A * D. nebras- 57.7 3.04 2.29 67.0 4.03 0.07 0.99 0.90 0.98 0.66 cense

Sc 1001B AMNH D. nebras- 35.9 3.02 2.18 41.0 4.10 0.11 – – – 0.59 3340 cense

USNM 547293 USGS B. scotti 37.6 – – 37.5 3.12 0.08 – – 0.80 – D84

USNM 547294 USGS B. scotti 54.4 – 2.06 62.5 – – – – – – D84

USNM 547295 USGS B. scotti 45.4 3.07 – 64.0 4.01 0.09 0.81 – – – D3941

USNM 547296 USGS B. scotti 67.3 3.24 2.21 37.0 5.00 0.07 0.99 1.03 1.24 0.78 D3464

microconch, from the Baculites scotti Zone, Pierre Shale, AMNH loc. 3386, Butte County, South Dakota ; and AMNH 81467 , a microconch, from the Baculites scotti Zone, Pierre Shale, AMNH loc. 3386, Butte County, South Dakota .

MATERIAL: Approximately 150 specimens of which 33 macroconchs and 70 microconchs comprise the measured set (tables 2 and 3). All of the specimens are from the Baculites scotti and overlying Didymoceras nebrascense zones of the Pierre Shale in South Dakota, Colorado, Montana, and Wyoming. All of the specimens are internal molds without any shell material.

MACROCONCH DESCRIPTION: In the measured sample, LMAX averages 42.7 mm and ranges from 27.9 to 79.8 mm (fig. 4, table 2). The ratio of the size of the largest specimen to that of the smallest is 2.86. The distribution is unimodal with a peak at 35–40 mm. However, in samples from single localities, the size range is much more restricted (fig. 5). At USGS Mesozoic loc. D3935, LMAX ranges from 35.2 to 40.4 mm and the ratio of the size of the largest specimen to that of the smallest is 1.15. At AMNH loc. 3386, LMAX ranges from 32.0 to 55.2 mm and the ratio of the size of the largest specimen to that of the smallest is 1.72.

Adults are slender, with a rounded to elongate outline in lateral view. For example, the outline is rounded in USNM 547292 ( LMAX /HS = 2.00), whereas it is elongate in AMNH 64601 ( LMAX /HS = 2.45). The exposed phragmocone occupies approximately one-half whorl. It terminates slightly below the line of maximum length. The umbilicus is small and deep, averaging 3.79 mm in diameter. The ratio of UD/ LMAX averages 0.09.

The body chamber consists of a short shaft and recurved hook, extending slightly beyond the coiled portion, leaving a small gap. LMAX / HP averages 3.13 and ranges from 2.91 to 3.27

TABLE 3 Measurements of Hoploscaphites gilberti , n. sp., microconchs See figure 3 for description of measurements. All measurements are in mm.

Specimen Locality Zone LMAX LMAX / UD UD/ WP/ WS/ WH/ VS/ Number HP LMAX HP HS HH HS AMNH 63525 AMNH 3386 B. scotti 37.1 3.34 3.96 0.11 – 0.78 1.02 –

AMNH 63528 AMNH 3386 B. scotti 33.3 3.74 3.47 0.10 1.10 0.88 1.08 – AMNH 63529 AMNH 3386 B. scotti 31.8 3.42 3.27 0.10 0.85 – – – AMNH 63532 AMNH 3386 B. scotti – – 4.02 – – – 0.97 –

AMNH 63534 AMNH 3386 B. scotti 32.0 3.11 4.06 0.13 1.02 0.92 – 0.71

AMNH 63535 AMNH 3386 B. scotti 26.8 – – – – 0.94 – 0.68

AMNH 63536 AMNH 3494 D. nebrascense 36.1 3.57 4.14 0.11 0.96 1.02 – 0.66

AMNH 63538 AMNH 3494 D. nebrascense 38.3 3.58 4.70 0.12 1.07 1.01 1.08 – AMNH 64493 AMNH 3386 B. scotti 39.8 – – – – – – –

AMNH 64549 AMNH 3386 B. scotti 36.0 3.75 3.81 0.10 0.93 – 1.06 –

AMNH 64573 AMNH 3386 B. scotti 34.7 3.44 4.28 0.12 0.88 1.01 1.04 0.73

AMNH 64575 AMNH 3386 B. scotti 30.4 3.01 2.91 0.10 – 0.96 0.97 0.74

AMNH 64589 AMNH 3386 B. scotti 35.8 3.44 – – 0.90 0.76 1.04 0.55 AMNH 64603 AMNH 3386 B. scotti 34.1 3.48 4.10 0.12 0.87 0.93 1.04 –

AMNH 64615 AMNH 3386 B. scotti 38.0 3.36 4.37 0.12 0.88 0.93 1.08 0.70

AMNH 64617 AMNH 3386 B. scotti – – 2.29 – 1.03 – – –

AMNH 64618 AMNH 3340 D. nebrascense 39.1 – 4.65 0.12 – – – –

AMNH 64631 AMNH 3386 B. scotti 34.2 3.26 3.55 0.10 0.93 0.90 0.96 0.61 AMNH 64645 AMNH 3386 B. scotti – – 2.30 – – 0.98 1.06 0.71 AMNH 64659 AMNH 3386 B. scotti 35.5 3.48 3.88 0.11 1.00 0.95 0.99 0.68

AMNH 64660 AMNH 3386 B. scotti 27.6 3.33 3.47 0.13 0.87 – – –

AMNH 64677 AMNH 3386 B. scotti 34.7 3.07 4.23 0.12 0.88 0.89 – –

AMNH 81094 AMNH 3340 D. nebrascense 36.3 – 3.56 0.10 – – – –

AMNH 81098 AMNH 3386 B. scotti 37.4 3.14 – – 0.81 – – – AMNH 81108 AMNH 3386 B. scotti 28.8 3.35 4.19 0.15 0.86 – – –

AMNH 81118 AMNH 3340 D. nebrascense 44.1 3.34 4.32 0.10 0.90 0.88 – –

AMNH 81467 AMNH 3386 B. scotti 35.3 3.57 3.49 0.10 0.92 0.97 1.10 –

AMNH 81451 AMNH 3386 B. scotti 34.3 3.59 – – 1.01 0.91 – – AMNH 81472 AMNH 3386 B. scotti 26.4 3.18 2.52 0.10 0.80 0.83 1.08 0.71 AMNH 81503 AMNH 3344 D. nebrascense 33.3 3.33 3.61 0.11 – 0.81 0.99 –

Sc 1001J AMNH 3340 D. nebrascense 37.6 – 3.64 0.10 – 0.89 – –

Sc 1289A AMNH 3340 D. nebrascense 40.0 3.74 4.14 0.10 0.93 0.84 1.05 0.57

USNM 28362 USGS 1359 B. scotti 34.0 3.24 2.56 0.08 0.87 0.93 1.07 0.75

USNM 547306 USGS D3935 B. scotti 28.1 3.16 3.01 0.11 0.82 0.98 – – USNM 547307 USGS D3935 B. scotti 23.8 3.35 3.07 0.13 0.89 0.89 0.94 0.75

USNM 547308 USGS D3935 B. scotti 36.0 3.43 – – 0.90 0.81 0.88 –

(3.23 in the holotype). The hook is weakly reflected, as indicated by a low apertural angle, averaging 52.1° and ranging from 32.5 to 74.0°. The apertural lip is flexuous with a deep constriction. In lateral view, the umbilical shoulder of the shaft is convex or straight, depending on the presence or absence of an umbilical swelling. The umbilical swelling appears as an elongate bulge that extends several millimeters along the shaft, as in USNM 547301 (fig. 7I–K).

The whorl section of the phragmocone along the line of maximum length is compressed. WP / HP averages 0.91 and ranges from 0.79 to 1.09 (0.92 in the holotype). The intercostal whorl section is subquadrate with maximum width at one-third whorl height. The umbilical wall is steep and subvertical and the umbilical shoulder is sharply rounded. The flanks are very broadly rounded to flat, converging toward the venter. The ventrolateral shoulder is sharply rounded and the venter is more broadly rounded .

Whorl width gradually increases as the shell passes into the shaft, and reaches its maximum value at the point of recurvature. Whorl height increases more abruptly and reaches its maximum value at midshaft, after which it decreases slightly. The whorl section at midshaft is slightly more compressed than that of the phragmocone along the line of maximum length. WS/HS averages 0.81 and ranges from 0.70 to 1.03 (0.83 in the holotype). The ratio of VS/HS averages 0.63 and ranges from 0.53 to 0.78 (0.65 in the holotype). The intercostal whorl section at midshaft is subquadrate. The umbilical shoulder is sharply rounded on the shaft, becoming more broadly rounded on the hook. The flanks are subparallel and very broadly rounded to nearly flat, with maximum width at one-third whorl height. In some specimens such as USNM 547294 (fig. 8F), the inner flanks are broadly rounded, and the outer flanks are flattened and converge toward the venter. The ventrolateral shoulder is sharply rounded and the venter is more broadly rounded.

The whorl section becomes less compressed from midshaft to the point of recurvature. WH/ HH averages 1.00 and ranges from 0.86 to 1.24 (1.02 in the holotype). The intercostal whorl section also becomes slightly more ovoid, with maximum width at one-half whorl height.

All specimens, regardless of their size, display the same pattern of ornamentation consisting of fine, flexuous ribs, umbilicolateral bullae, and ventrolateral tubercles, which usually extend to the aperture. Although this pattern of ornamentation is the same among all specimens, the spacing of the ribs and the size of the umbilicolateral bullae and ventrolateral tubercles vary depending on adult size.

Starting at the point of exposure, primary ribs arise at the umbilical seam and strengthen across the umbilical wall and shoulder. They are rectiradiate to prorsiradiate on the flanks. They are straight or weakly flexuous, bending slightly backward on the inner flanks, slightly forward on the midflanks, and slightly backward again on the outer flanks. They are sharp, narrow, and equally strong on the venter, which they cross with a slight adoral projection. The rib density on the phragmocone along the line of maximum length ranges from 7 to 14 ribs/cm (11 ribs/ cm on the holotype). Branching and intercalation occur at one-third and two-thirds whorl height, at the umbilicolateral and ventrolateral margin, respectively.

On the body chamber, ribs arise at the umbilical seam and are rursiradiate on the umbilical wall and shoulder, becoming more rectiradiate toward the adoral end of the shaft. In larger, more coarsely ornamented specimens, such as USNM 547304 (fig. 8A–C), the ribs on the umbilical wall and shoulder are raised into barlike bullae. Ribs weaken on the flanks; they are rectiradiate to slightly rursiradiate on the adapical portion of the shaft, becoming more prorsiradiate on the adoral portion of the shaft. Ribs are narrow and slightly flexuous, swinging gently backward on the inner flanks, more strongly forward on the midflanks, and weakly backward again on the outer flanks. This flexuosity diminishes on the hook. Branching and intercalation occur at one-third and two-thirds whorl height, at the umbilicolateral and ventrolateral margin, respectively. Ribs cross the venter with a marked forward projection. The rib density at midshaft ranges from 8 to 13 ribs/cm (10 ribs/cm on the holotype). The rib density remains the same or increases on the hook (17 ribs/cm on the holotype).

In most small specimens, such as the holotype, umbilicolateral bullae are absent on the phragmocone. Instead, the primary ribs in this area are strong and adorally concave. In larger specimens, umbilicolateral bullae appear midway or near the adoral end of the exposed phragmocone. For example, four umbilicolateral bullae are present on the adoral portion of the exposed phragmocone in USNM 547296 (fig. 9A–C). Bullae are relatively evenly spaced, with a maximum distance between consecutive bullae of 5 mm. As many as two ribs intercalate between bullae; one rib joins a bulla dorsally and two or three ribs branch from it ventrally .

Umbilicolateral bullae are usually, but not invariably, present on the body chamber. If bullae are absent, as in the holotype, the primary ribs in this area are strong and adorally concave. If bullae are present, they occur on the primary ribs at one-third whorl height. In some specimens, such as USNM 547302 , the bullae are elevated into tubercles 1.5 mm high (fig. 6Y, Z). The maximum total number of umbilicolateral bullae on the body chamber is 8 ( USNM 547294 ), and the maximum total number on the entire shell is 13 ( USNM 547304 ). The bullae are more or less evenly spaced on the adapical part of the shaft, becoming slightly more widely spaced on the adoral part, where they attain their maximum height. They subsequently diminish in size and spacing toward the aperture. In the largest specimen, USNM 547296 , the maximum distance between consecutive bullae on the body chamber is 9 mm (fig. 9A–C). One or two ribs join a bulla dorsally, and two to four ribs branch from it ventrally. As many as two ribs intercalate between bullae .

Ventrolateral tubercles appear on the shell anywhere from the point of exposure, as in USNM 547298 (fig. 6D, E), to the adoral end of the phragmocone, as in USNM 547295 (fig. 7C–E).Very rarely are they absent altogether on the phragmocone, as in USNM 547301 (fig. 7I–K). They are relatively small on the phragmocone, less than 1.5 mm in height. In general, they are unevenly spaced with gaps between successive tubercles. For example, in USNM 547295 (fig. 7C–E), starting at the point of exposure, the distance between successive tubercles, as measured in an adoral direction, is 3, 3, 3, 5.5, 3, 6, and 7.5 mm. The distribution is also uneven in USNM 547300 (fig. 7P, Q). Starting at the point of exposure, the distance between successive tubercles, as measured in an adoral direction, is 5, 5, 6, 6.5, and 4.5 mm. More rarely, the tubercles on the phragmocone become gradually more widely spaced in an adoral direction. For example, in USNM 547296 (fig. 9A–C), starting midway on the exposed phragmocone, the distance between successive tubercles gradually increases, attaining a maximum value of 6.5 mm at the adoral end of the phragmocone. The maximum total number of ventrolateral tubercles on the exposed phragmocone is 12. One or two ribs join a ventrolateral tubercle dorsally and two or three ribs branch from it ventrally, looping to ventrolateral tubercles on the opposite side of the venter. As many as five ribs intercalate between ventrolateral tubercles, depending on the distance between tubercles and the density of ribbing.

Ventrolateral tubercles are invariably present on the body chamber and occur at 85%–90% whorl height. In a few specimens, such as the holotype, the largest gap between ventrolateral tubercles occurs between the most adoral tubercle on the phragmocone and the most adapical tubercle on the body chamber. Thereafter, the tubercles are nearly always more or less uniformly spaced. For example, in USNM 547302 (fig. 6Y, Z), starting at the adapical end of the body chamber, the distance between successive tubercles, as measured in an adoral direction, is nearly constant (6.5, 5, 5, 5.5, 5.5, and 6 mm). The distribution of ventrolateral tubercles on the body chamber is also nearly uniform in USNM 547293 (fig 6Q, R). Starting at the adapical end of the body chamber, the distance between successive tubercles, as measured in an adoral direction, is 4.5, 4.5, 4, 5, 5, 3.5, and 5 mm .

Ventrolateral tubercles attain their maximum height at midshaft (1.5 mm). In some specimens, the tubercles become weakly clavate at this point, with a flat, steepened adapical face, and a more gently sloping adoral face. In smaller, more finely ornamented specimens such as the holotype, the ventrolateral tubercles die out on the adoral end of the shaft. In larger, more coarsely ornamented specimens such as USNM 547294 (fig. 8F), the ventrolateral tubercles persist onto the hook, where they become smaller, more bullate, and more closely spaced. The total number of ventrolateral tubercles on the body chamber ranges from 6 to 14 (8 in the holotype). The total number of ventrolateral tubercles on the entire adult shell ranges from 6 to 21 (8 in the holotype). At midshaft, one to three ribs join a ventrolateral tubercle dorsally and two or three ribs branch from it ventrally, looping between ventrolateral tubercles on the opposite side of the venter. Two to four ribs intercalate between ventrolateral tubercles, depending on the distance between tubercles and the density of ribbing .

The suture is simple, with a broad, asymmetrically bifid first lateral saddle (E/L) and a narrow, symmetrically to asymmetrically bifid first lateral lobe (L) (fig. 12).

MICROCONCH DESCRIPTION: LMAX averages 32.9 mm and ranges from 23.8 to 44.1 mm (table 3). The size distribution is unimodal, with a peak at 30–35 mm (fig. 4). The ratio of the size of the largest specimen to that of the smallest is 1.85. The size distribution of macroconchs and microconchs overlaps. The average size of microconchs is 77% that of macroconchs. The size range of microconchs is reduced in samples from single localities (fig. 5). At USGS Mesozoic loc. D3935, LMAX ranges from 23.8 to 36.0 mm; the ratio of the size of the largest specimen to that of the smallest is 1.51. At AMNH loc. 3386, LMAX ranges from 26.4 to 39.8 mm; the ratio of the size of the largest specimen to that of the smallest is also 1.51.

Shells range from circular to elongate in lateral view, clustering more heavily on the elongate end of the spectrum. The exposed phragmocone occupies one-half to two-thirds of a whorl. In general, the base of the body chamber occurs below the line of maximum length. The umbilicus of the phragmocone is relatively small. UD average 3.60 mm and UD/LMAX averages 0.11. The body chamber consists of a relatively long shaft and recurved hook, leaving a small gap between the exposed phragmocone and hook. The ratio of LMAX/HP averages 3.33, which is greater than that in macroconchs (3.13), implying that microconchs are relatively more uncoiled than macroconchs. The apertural lip is flexuous, with a short dorsal projection. The umbilical seam of the shaft is concave in lateral view revealing the umbilicus of the phragmocone. The curvature of the seam closely matches that of the venter, epitomizing the morphology of microconchs.

The intercostal whorl section of the phragmocone along the line of maximum length is compressed subquadrate. WP/HP averages 0.90 (0.90 also in macroconchs). The umbilical wall is steep and subvertical and the umbilical shoulder is sharply rounded. The inner flanks are broadly rounded and the outer flanks are nearly flat and converge toward the venter. Maximum width occurs at one-third whorl height. The ventrolateral margin is sharply rounded and the venter is more broadly rounded.

Whorl width and height gradually increase as the shell passes into the shaft. Whorl width attains its maximum value at the point of recurvature, and whorl height attains its maximum value at midshaft. The increase in whorl height from the phragmocone to the shaft is much more gradual than that in macroconchs.

The whorl section at midshaft exhibits nearly the same degree of compression as that of the phragmocone along the line of maximum length (WS/HS averages 0.90). The ratio of VS/HS averages 0.68 and ranges from 0.55 to 0.83.The intercostal whorl section is subquadrate with maximum width at one-quarter whorl height, corresponding to the site of the umbilicolateral bullae. The umbilical wall is steep and the umbilical shoulder, unlike that in macroconchs, slopes gently outward. The flanks are broadly rounded to flat and converge toward the venter. The ventrolateral shoulder is sharply rounded and the venter is more broadly rounded.

As in macroconchs, the whorl section becomes slightly less compressed from midshaft to the point of recurvature, after which it remains nearly the same. WH/HH averages 1.02. The intercostal whorl section at the point of recurvature is slightly more ovoid than that at midshaft. The umbilical wall is nearly flat and the umbilical shoulder is sharply rounded. The flanks are broadly rounded, with maximum width at one-third whorl height. The ventrolateral shoulder is sharply rounded and the venter is more broadly rounded.

The ornament in microconchs is similar to that in macroconchs. Specimens are covered with relatively fine, flexuous ribs, and rows of umbilicolateral bullae and ventrolateral tubercles. However, in many small specimens, these latter features only appear on the shaft of the body chamber.

At the point of exposure, ribs are rursiradiate on the umbilical wall and shoulder. They are prorsiradiate and slightly flexuous on the flanks, swinging slightly backward on the inner flanks, slightly forward on the midflanks, and slightly backward again on the outer flanks, crossing the venter with a slight forward projection. Branching and intercalation occur at onethird and two-thirds whorl height, at the umbilicolateral and ventrolateral margin, respectively. The rib density on the phragmocone along the line of maximum length ranges from 9 to 11 ribs/cm except in very small specimens such as USNM 547326 (fig. 10V–Y) in which the density is as high as 16 ribs/cm.

On the body chamber, primary ribs arise at the umbilical seam and are strong on the umbilical wall. They are narrow and weakly flexuous on the flanks, bending slightly backward on the inner flanks, slightly forward on the midflanks, and slightly backward again on the outer flanks. Ribs are rectiradiate on the adapical end of the shaft, becoming more prorsiradiate on the adoral end of the shaft. Branching and intercalation occur at the umbilicolateral margin (at the bullae if they are present) and at the ventrolateral margin (at and between the ventrolateral tubercles). Ribs cross the venter with a weak adoral projection. The rib density at midshaft ranges from 12 to 16 ribs/cm, with the exception of 22 ribs/cm in USNM 547326 (fig. 10V–Y), and remains the same or increases slightly on the hook.

Umbilicolateral bullae are absent on the phragmocone in small specimens and, instead, the ribs in this area are strong and adorally concave. Even in large specimens, as in USNM 547324 (fig. 10cc–ee), umbilicolateral bullae appear only on the adoral one-third of the phragmocone. They occur at one-third whorl height and are linked to the umbilicus by strong primary ribs. The bullae are more or less evenly spaced with a maximum distance of 3 mm between them. One rib joins a bulla dorsally, and two ribs branch from it ventrally. One or two ribs intercalate between bullae.

Umbilicolateral bullae are always present on the body chamber, with the exception of very small specimens such as USNM 547314 (fig. 10G–I). The maximum total number of umbilicolateral bullae on the body chamber is 6, and the maximum total number of bullae on the adult shell is 9. The bullae occur at one-quarter whorl height on the shaft, but migrate closer to the umbilical shoulder on the hook. They are small and attain their maximum height (1.5 mm) on the adoral end of the shaft. They are more or less evenly spaced, with some approximation near the aperture. For example, in USNM 547324 (fig. 10cc-ee), the bullae are nearly uniformly spaced at intervals of 3–4 mm. At midshaft, one rib joins an umbilicolateral bulla dorsally and two or three ribs branch from it ventrally. One or two ribs intercalate between successive bullae.

Ventrolateral tubercles are present on the exposed phragmocone in nearly all of the specimens in our sample. The maximum total number of tubercles on the exposed phragmocone is 9, but such a high number is rare, with most specimens exhibiting only 2 to 7 tubercles. In general, tubercles appear between the middle and adoral end of the phragmocone. They are unevenly spaced with gaps between them. For example, in AMNH 81467 (fig. 10ff–hh), starting near the point of exposure, the distance between successive tubercles, as measured in an adoral direction, is 2, 2.5, 2.5, 5.5, 3, and 2.5 mm. Commonly, one rib joins a ventrolateral tubercle dorsally and two ribs branch from it ventrally, looping to ventrolateral tubercles on the opposite side of the venter, with one to three intercalated ribs between them.

Ventrolateral tubercles are invariably present on the shaft, but rarely continue onto the hook. They are less than 1.5 mm in height, and are smaller than the adjacent umbilicolateral bullae. They occur at 90% whorl height and are paired or offset from one side of the venter to the other. They are more or less uniformly spaced although gaps are not uncommon, especially at the adapical and adoral ends of the shaft. For example, in USNM 547324 (fig. 10cc–ee), the distance between consecutive tubercles, as measured in an adoral direction, is 4, 3.5, 3.5, 3.5, 4.5, 3, 5.5, and 8.5 mm. The total number of ventrolateral tubercles on the body chamber ranges from 5 to 12, and the total number of tubercles on the entire adult shell ranges from 9 to 16. One to three ribs join a ventrolateral tubercle dorsally and two to four ribs branch from it ventrally. Three to six ribs intercalate between tubercles, depending on the density of ribbing and the spacing between tubercles.

The suture of microconchs is similar to that of macroconchs (fig. 12).

JAWS AND HOOKLIKE STRUCTURES: AMNH locs. 3340 and 3386 contain abundant jaws and hooklike structures (fig. 13). The jaws appear as isolated occurrences and are attributed to Hoploscaphites gilberti , n. sp., or an as yet undescribed, more coarsely ornamented species of Hoploscaphites . The lower jaw conforms to the structural plan of the Aptychophora ( Landman et al., 2010). It consists of two symmetrical wings separated by a slit down the middle (fig. 13A). The upper jaw consists of a pair of widely open inner lamellae and a short, reduced outer lamella, both of which converge toward the apex (fig. 13B).

The hooklike structures are approximately 2 mm in maximum length and are composed of the same black material as the jaws, presumably originally chitin (fig. 13C–E). The most complete specimens consist of two hollow projecting points extending from a bulbous base. These structures have been documented in other species of Hoploscaphites by Landman and Waage (1993). Similar but much larger structures have also been described in Rhaeboceras by Kennedy et al. (2002). The function of these structures is unknown.

DISCUSSION: In his description of the Pierre Shale of Colorado, Gilbert (1896: pl. 63, fig. 3) illustrated a specimen of what he called a small variety of Scaphites nodosus (USNM 28362). It is from a limestone core of a tepee butte in the Pierre Shale at USGS loc. 1359, Pueblo County, Colorado, and a microconch of Hoploscaphites gilberti , n. sp. (fig. 11W–Y). Gilbert (1896: pl. 65, fig. 2) also illustrated a specimen of what he called a large individual of S. nodosus (USNM 28363). It is from the Hygiene Sandstone Member of the Pierre Shale of Colorado (what he called the “Tepee zone of the Pierre Shale”). This specimen is a macroconch of an as yet undescribed closely related species of H. nodosus from the Baculites reesidei Zone.

Hoploscaphites gilberti , n. sp., most closely resembles H. gilli Cobban and Jeletzky, 1965 , which ranges from the Baculites perplexus to Didymoceras stevensoni zones of the Western Interior of North America (middle middle to lower upper Campanian). Indeed, Cobban and Jeletzky (1965: pl. 95, fig. 1A–D) illustrated what they described as a coarsely ornamented specimen of H. gilli from southeastern Montana on the northwestern flanks of the Black Hills, which is actually a microconch of H. gilberti , n. sp. Several morphological features distinguish H. gilberti , n. sp., from H. gilli (fig. 14). In terms of the whorl cross section, the flanks of the body chamber are nearly subparallel in H. gilberti , n. sp., whereas they are more steeply convergent toward the venter in H. gilli . In terms of ornamentation, the two species differ in the density of ribbing, with H. gilberti , n. sp., having more widely spaced ribs. For example, the density of ribbing on the midshaft of the holotype of H. gilberti , n. sp., is 10 ribs/cm whereas it is 13 ribs/cm on the midshaft of a macroconch of H. gilli of comparable size (fig. 14D–F). In addition, H. gilberti , n. sp., bears umbilicolateral bullae on the body chamber whereas these features are absent in H. gilli . In contrast, ventrolateral tubercles are present on the body chamber of both species, although they are larger, more numerous, and more closely spaced in H. gilberti , n. sp., than in H. gilli .

Hoploscaphites gilberti , n. sp., is also similar to H. brevis Meek, 1876 , which ranges from the Didymoceras cheyennense to Baculites cuneatus zones of the Western Interior of North America (middle upper Campanian). However, the venter at midshaft is more rounded in H. gilberti , n. sp., than in H. brevis . As a result, the distance between ventrolateral tubercles on either side of the venter is greater in H. gilberti , n. sp., than in H. brevis . For example, the value of Vs/Hs, which is a relative measure of the distance between ventrolateral tubercles on either side of the venter, averages 0.62 in our sample of 14 macroconchs of H. gilberti , n. sp., versus 0.43 in a sample of 54 macroconchs of H. brevis , as recorded in Landman et al. (2010: table 7).

Hoploscaphites gilberti , n. sp., is strongly dimorphic, with the dimorphs referred to as macroconchs (figs. 6–9) and microconchs (figs. 10, 11). The dimorphs are interpreted as sexual in nature, the macroconchs female, and the microconchs male, following the traditional view as outlined by Makowski (1962) and Cobban (1969). Dimorphs are distinguished at maturity by the presence of an umbilical bulge, the outline of the umbilical shoulder relative to that of the venter in side view, the size of the umbilical diameter, the change in whorl height from the mature phragmocone to the midshaft of the body chamber, and adult size.

Some of these criteria are more reliable than others in distinguishing between dimorphs. For example, the size of the largest macroconch is larger than the size of the largest microconch, but the size distributions of the two dimorphs overlap. In addition, the curvature of the umbilical seam is not an infallible indicator of the dimorph. In some specimens, the umbilical shoulder is straight, suggesting a macroconch, but other features of the shell, including a gradual rather than abrupt increase in whorl height in passing from the phragmocone to the midshaft, suggest a microconch instead. Landman et al. (2010) called such specimens “microconchs, transitional to macroconchs.” In our sample of Hoplocaphites gilberti , n. sp., we classify AMNH 81100 (fig. 6I–K) as a macroconch because it exhibits a straight umbilical shoulder. However, the increase in whorl height in passing from the phragmocone to the midshaft is less abrupt than that in most macroconchs, and, as a result, the shell is more elongate.

Hoploscaphites gilberti , n. sp., shows a broad range in adult size. In macroconchs, LMAX ranges from 27.9 to 79.8 mm, and, in microconchs, it ranges from 23.8 to 39.8 mm. Most of our specimens are from cold methane seep deposits in the Baculites scotti and overlying Didymoceras nebrascense zones of the Pierre Shale. However, three of the largest specimens, USNM 547304 (fig. 8A–C), 547296 (fig. 9A–C), and Sc 590A (fig. 8D, E) are from age-equivalent nonseep sites. It is possible that the variation in adult size between sites may reflect differences in the timing of maturation. Assuming that all specimens grew at approximately the same rate, the smaller size of the specimens from the seeps may indicate that they matured at a smaller size due to the abundant supply of nutrients at these sites.

OCCURRENCE: This species occurs in the Baculites scotti and overlying Didymoceras nebrascense zones (upper middle to lower upper Campanian) of the U.S. Western Interior. It is especially common in the tepee buttes along the Front Range of Colorado and the Black Hills region of South Dakota, Wyoming, and Montana. These sites represent deposits associated with cold methane seeps ( Gilbert and Gulliver, 1895; Kauffman et al., 1996; Landman et al., 2012).

Figures

FIG. 4. Size-frequency histogram of Hoploscaphites gilberti, n. sp., Pierre Shale, Baculites scotti–Didymoceras nebrascense zones, based on the samples in tables 1 and 2.