Research Spotlight: November 2018

Tue, 2018-11-13 14:04 -- maphillips

Augustus Fendler Herbarium Specimens: A Locality Improvement Project

A component of the Southern Rocky Mountain Flora Database Project
Lance J. Gloss and Timothy J. S. Whitfeld
Brown U. Herbarium (BRU)
Dec 2017 - May 2018

Introduction

This project involved archival research into primary and secondary sources to improve the precision and accuracy of our records for the "Plantae Fendlerianae"collection made by Augustus Fendler (1813-1883) on an 1846-1847 expedition to the area “Novo Mexicanae.” By comparing sets of field notes to the collecting numbers and species names in the physical archive, we drastically improved the locality precision for ~393 of Fendler’s specimens.

The Augustus Fendler “Novo-Mexicanae” Specimens

The Brown University Herbarium has been engaged for just over four years in digitizing the archive of physical plant specimens. Our digitizing team at Brown had previously added all of our existing Fendler specimens to the herbarium database (Specify v.6.6.06). As of January 2017, the typical Fendler specimen had only a vague, sparse record. Generally, the information was as follows: the collector’s name “A. Fendler”; the date 1847; the locality “New Mexico”; the locality note “Plantae Novo-Mexicanae” as it appears on > 95% of the specimens; a species or genus determination (either the original, or any updated name attached to the specimen since 1847); and a collection number. The "Plantae Fendlerianae" collection at BRU is itself incomplete. Our 393 specimens included a range of collection numbers between 1-1024.

The paucity of information on each of the Fendler specimens is not atypical of herbarium specimens from that period. In this case, we were aware by 2016 that our locality information for this collection was particularly imprecise. This is because the territory referred to as New Mexico in 1847—appearing in the collection title as Novo-Mexicanae—covers a territory including the majority of present-day Arizona and New Mexico (minus the Gadsden Purchase Territory) as well as parts of Colorado and Nevada. Furthermore, the collection dates coincide precisely with the Mexican-American War, which caused serious turmoil in the region. To georeference based on the locality indicator "Novo-Mexicanae" alone would have led either to a) the misrepresentation of this location as the centroid of the contemporary US state of New Mexico, or b) a reference point near today’s Arizona-New Mexico border, with a margin of error as great as ~300 miles (~500 km).

In an effort to improve on these prospects for georeferencing, we turned to primary and secondary sources about August Fendler’s trip. His journey resembled many botanical expeditions in the western United States during the 19th century. Elizabeth A. Shaw of the Gray Herbarium at Harvard University describes the circumstances of the 1846-1847 Fendler expedition in detail. Augustus Fendler had commenced his studies of natural history as a young man, not long after immigrating from Germany to the United States. By the mid-1840s, he was an accomplished collector, and had developed close relationships with leading botanists, including the independent collector George Engelmann (also a recent immigrant from Germany) and Asa Gray from Harvard University, the preeminent botanist of the time. When Asa Gray caught wind of a US army mission to the southwest and California, Engelmann suggested to Gray that Fendler accompany the soldiers as the expedition's botanist. Under these circumstances, Fendler and this so-called "Army of the West" embarked.

We now know that on August 10, 1846, Fendler departed Fort Leavenworth, Kansas, and travelled along the Arkansas River through Kansas and Southern Colorado, past Bent’s Fort. He reached northern (present-day) New Mexico on September 29, 1846, and by October 11 he had reached Santa Fe, the primary collecting site. For the following ten months, he made collections in that vicinity, which make up some 85% of the Fendler specimens at BRU. On August 9, 1847, he and his company left the mountains around Santa Fe and returned to Fort Leavenworth on a route further to the south and east than the one on which they had came, crossing the Cimarron River at the far western end of the Oklahoma panhandle, and arriving at the Fort in Kansas on September 24, 1847.

Locality Improvement Method

Because we had the collection number for each specimen, we postulated that our data--both date and locality--could be improved with the help of secondary sources, as long as any sources also had the corresponding numbers. A search of historical and botanical archives online yielded three publications that together enabled this project.

The first source we found on Fendler’s trip was The Type Localities of Plants First Described From New Mexico (Standley 1910). The document included several Fendler specimens, but these were all types, for which information is readily available. The Standley publication did, however, describe Fendler’s route, and was the first to make apparent the arc of his expedition. This, we learned, did not include present-day Arizona, then part of the New Mexico territory, but did include areas of present-day Colorado, Oklahoma, and Kansas.

The next publication contained a large portion of the information we were looking for: Plantae Fendlerianae Novo Mexicanae: An account of a collection of plants made chiefly in the vicinity of Santa Fe, New Mexico by Augustus Fendler, an 1848 book by Asa Gray. This publication yielded information for 462 specimens, which were catalogued by number and species name. Unfortunately, it appears that after authoring the first half of this review in 1848, Gray did not fulfill the “to be continued” note at the end of the book. This record therefore only allowed the improvement of ~50% of our records. Furthermore, we found that not all of the numbered and named specimens in our collection correlated to the number-name combinations in the document.

The third and most useful publication, Augustus Fendler's Collection List: New Mexico, 1846-1847 (Shaw 1982), cleared up the issue of mismatched names and numbers. Shaw published this review of primary sources related to Fendler at the Gray Herbarium. Her paper confirmed and elaborated the localities of the 1910 Standley document. It also lists every specimen collected by Fendler, with the collection number used by Gray and a second set of numbers that Shaw attributes to Fendler and his fellow collector, George Engelmann. This explained the apparent multiplicity of numbering schemes. In many respects, Shaw’s text was the key to this process. However, some limitations of her paper made the other documents necessary. For example, Shaw grouped specimens by genus, species, but we could overcome this relatively easily through a comparison of the collection numbers. Furthermore, Shaw’s locality notes were generally less detailed than Gray’s, so--when available--Gray’s locality notes were the ones added to the database.

A handful of the new locality entries reflect compromises made to accommodate some apparent contradictions of different sources. Shaw notes that when Gray recorded the specimens and grouped them by number, he combined specimens of the same species, even if they were collected on different dates and in different localities. This is now reflected in our database, where the primary locality listed is the New Mexico locality for that plant (usually Santa Fe county), and any other localities that might be the actual origin of the specimen are listed as secondary notes. Another one of the persistent quality issues with the locality data relates to the disappearance of sites referenced by Fendler from the lexicon and from the landscape itself. Many 19th century geographical references persist, but for others we can have little hope of precision. River crossings, such as the “Cimarron River Ford” mentioned by Fendler in connection to two discrete locations along his route, is an example of such a challenge to contemporary georeferencing efforts. Where possible, we have resolved these issues through the support of previous authors writing on Fendler, particularly Shaw, when they have managed to triangulate the general location of places like the Cimarron Fords by closely reviewing the routes of Fendler and other contemporary travelers in the region.

By refining the localities of Fendler’s collections, based on these three publications, mean georeferencing error decreased from ~350 km to ~60 km. More precise date and habitat data have also been added to these specimen records. Similar improvements in precision were made in the other updates described below. In Wyoming, for example, mean georeferencing error improved from ~275 km to ~50 km.

Fendler Update Example

Specimen: PBRU00034131 (available on the portal of the Consortium of Northeastern Herbaria: www.neherbaria.org)

Species: Dalea alopecuroides

Collector Number on Sheet: 127

Writing on Sheet: “Plantae Novo-Mexicanae / No. 127 / 1847 / Herbarium Olneyanum  / Dalea alopecuroides Willd.”

Gray’s Entry: “127. Dalea alopecuroides, Willd. ; DC. Prodr. 2. p. 21. Low prairies, &c., Santa Fe ; also on the Arkansas River ; July to September.”

Shaw’s Entry: “141 (127) 10 July-15 September, 1846-1847. n.b.: 15 September, 1846 Santa Fe, margin of irrigating ditches, near fields. Also: low prairies near the Arkansas River.”

Current Database Entry: Santa Fe County, NM / 07/10/1847 - 09/15/1847 / No. 127 / A. Fendler / Plantae Novo-Mexicanae. Low prairies, Santa Fe, margin of irrigating ditches, near fields. Also: low prairies near the Arkansas River, Colorado.

Further Applications of Method

Given our success updating the Fendler specimens, we have attempted similar work on other sets of specimens with vague locality information and imprecise dates. Here, we describe two other examples.

The Charles C. Parry Wyoming Collection

The first of these is the 1873 collection made by Charles Parry in the state of Wyoming. BRU has ~30 specimens from Parry’s Wyoming collection, the locality information of which we have improved by a method closely paralleling the process described above. Parry also made extensive collections in Wyoming (~477 specimens at BRU), Utah (~325), California (~3), and elsewhere.

Like Fendler in New Mexico, Parry was accompanied to Wyoming by a contingent of soldiers, as well as several other scientists charged with preparing astronomical, geological, hydrological, and entomological reports. Parry authored the official botanical report for the expedition. This chapter in the 1875 Report upon the reconnaissance of northwestern Wyoming, including Yellowstone national park, made in the summer of 1873, was key to our endeavor to improve herbarium locality data for the Parry collection (Jones et al. 1875). The information contained in the 1875 report was then cross-referenced with a second publication, also authored by Parry, entitled Botanical Observations in Western Wyoming (1874). These entries are arranged by geography in the style of a journal, rather than as a systematized list of species in the collection. As such, some species that were collected are not mentioned at all in this document, while others are described in narrativized detail. Those species that are mentioned in the 1874 book have benefited from this additional detail about the localities in which he found them. Together, the two documents permitted an increase precision of the locality data.

Parry Update Example

Specimen: PBRU00049186

Species: Antennaria racemosa

Collector Number on Sheet: 177

Writing on Sheet: “Western N. American Botany / North-western Wyoming Expedition, Capt. W. A. Jones, U. S. Engineers, Commanding / No. 177 / Antennaria racemosa, Hook. / C. C. Parry, Coll. / 1873”

Entry in Parry’s 1874 Botanical Observations: “No. 248. Polygonum imbricatum, Nutt., Stinkingwater, July”

Entry in Parry’s chapter in the 1875 Report: “On reaching the upper portion of this [the Stinking Water, i.e. Shoshone River] valley, becoming more densely wooded, and frequently spreading out into open, grassy parks, a much more attractive and varied flora is brought to view. The pine woods, composed almost exclusively of Pinus contorta,  with scattering trees of Abies grandis... Here too occurs abundantly Antennaria racemosa Hook., with sterile and fertile plants growing in distinct plots...”

Current Database Entry: Park County, WY / 07/1873 / No. 177 / C. C. Parry / Stinkingwater. Upper Shoshone River Valley. North-western Wyoming Expedition, Capt. W. A. Jones, U.S. Engineers, Commanding.

The Daniel C. Eaton Utah Collection

Daniel C. Eaton was a student and later a colleague of Asa Gray’s. He participated in several scientific expeditions in the western US in the mid 19th-century, and went on to be recognized for his work on ferns, and as director of the Peabody Herbarium at Yale University. BRU holds ~128 specimens from a collecting expedition made by Eaton in Utah.

From 1867-1869, Eaton accompanied soldiers and scientists on the United States Geological Survey of the 40th Parallel. Their route along the 40th parallel took them to the mountain ranges around Salt Lake City, Utah, and further west of the Great Salt Lake Desert, to the mountains of the Great Basin in the what is today the state of Nevada. Another botanist, Sereno Watson, also collected on this journey. Watson accompanied the larger contingent of soldiers and surveyors into Nevada, while Eaton made his collections with a smaller detachment of soldiers in northern Utah.

The Geological Survey released a finding of it’s reports. Volume 5 of The United States Geological Exploration of the Fortieth Parallel (King et al. 1871) includes a chapter authored by Sereno Wilson, with Daniel C. Eaton and others, on the Botany of the region. Though the authors generally do not note which collections were made by Watson and which by Eaton, our knowledge of the expedition makes it clear that collections in the mountains of Nevada were made by Watson, while Eaton was responsible for collections made in the vicinity of Salt Lake City. Notes on a dozen entries support this interpretation. For example, a Poa specimen notes that it was “found by Prof. Daniel C. Eaton in a rocky gulch of Cottonwood Cañon in the Wahsatch.” Generally, the notes in this chapter have greatly improved the precision of our locality data.

Eaton Update Example

Specimen: PBRU00042054

Species: Eriogonum cernuum

Collector Number on Sheet: 888

Writing on Sheet:Eriogonum cernuum, Nutt., var. tenue, T. & G. / Utah, Jun.--Jul., 1869, coll. D. C. Eaton / Brown University Herbarium”

Entry in chapter on “Botany” by Watson & Eaton (1875): “... Western Texas to Arizona and north to Wyoming and Idaho. Var. tenue, T. & G. … Ruby Valley and Humboldt Pass, Nevada, on the foot-hills of the Wahsatch and in Bear River Valley, near Evanston, Utah; 5-6,000 feet altitude; July-September. No. 1,036.”

Current Database Entry: Park County, WY / 06/1869 - 17/1869 / No. 888 / D. C. Eaton / On the foothills of the Wasatch and in Bear River Valley, near Evanston, 5-6000 feet altitude. Geological Exploration of the Fortieth Parallel.

The Hall & Harbour Rocky Mountains Collection

One of the first major collections of flora from the Rocky Mountains was made by Elihu Hall and J. P. Harbour in 1862. BRU holds ~814 specimens from this collection, entitled "Rocky Mountain Flora, Lat 39-41". These specimens are exemplary of the challenges of georeferencing historical specimens with little data. Here, just a few small pieces of information that have enabled us to decrease mean georeferencing error for these specimens from ~370 km to ~85 km.

Elihu Hall and J. P. Harbour were botanists based in Illinois. Little is known about Harbour--no birth or death dates, academic appointments, or given names have been confirmed. By contrast, sources confirm that Elihu Hall lived in Illinois for most of his life (1822-1882). As a botanical collector on public and private contracts, Hall collected in numerous locations throughout the American Plains, Midwest, and Northwest, with his most notable work being the collections made in Colorado (1862), Oregon (1871), and Texas (1872).

The 1862 Hall &Harbour Collection poses numerous challenges to georeferencing. The labels consistently list the same limited information: "Rocky Mountain Flora, Lat 39-41, E. Hall & J.P. Harbour, Colls. 1862". Typically a specimen number is also listed.

This information has serious limitations. None of the Hall & Harbour specimens at BRU have a more specific location than the wide latitude band (39-41 deg. north) and the Rocky Mountains. The only exceptions are those specimens labelled “Alpine”, “Subalpine”, or “Alpine and Subalpine,” which offer vague clues toward determining elevation. Furthermore, all of the specimens are merely dated 1862, and collection numbers reputedly do not correspond to the sequence of collection.

A review of literature on the itinerary yields some clarifications. The collector Charles C. Parry had visited the mountains of Central Colorado in 1861, and had returned from that trip with a large collection labelled "Rocky Mountain Flora: From the headwaters of Clear Creek, and the alpine ridges lying east of ‘Middle Park’". Writings of that era by Parry (1868) and Asa Gray (1868), as well as modern sources such as Don Despain (2007), indicate that Parry returned on his second trip to more or less the same sites, this time accompanied by Hall & Harbour. Parry himself writes that the area under study in 1862 was bounded in the south by Pike’s Peak, and in the north by Clear Creek--roughly the area of Park County, Colorado.

Final clarity has been achieved through a comparison of existing sources. The botanist William Weber, who in 1961 confronted the question of Hall & Harbour’s itinerary, found “incidentally” that a report by the botanist George Engelmann included a series of barometric measurements made by Parry on his trip with Hall & Harbour. These measurements make their route quite clear. After arriving in Denver from Omaha, Nebraska, the party ascended into the Rocky Mountains at Mount Vernon, 20 km west of Denver. The party then traveled throughout South Park (a wide grassland valley roughly equivalent to Park County), reaching the town of Tarryall in southern Park County. From there, they ventured to Pike’s Peak in El Paso County, which they ascended on July 1st, 1862. The route thereafter was a loop back northward, through Colorado City (present-day Colorado Springs, El Paso County), Denver (through Jefferson County), Empire (Clear Creek County), and Middle Park (Grand County), leading them to finish the expedition at Gray’s Peak (Park County).

The entirety of this route lies within Park County or within 30km of the Park County border. For this reason, Park County serves as a far-improved locality for georeferencing of these specimens. Furthermore, this locality improvement project, in particular, may assist in locality improvement for several related collections.

Namely, ~20 specimens are labeled "Rocky Mountain Flora Lat 39-41" but have a different label, and are credited to Charles C. Parry, 1862. This is consistent with the literature, which says Parry collected several plants on the trip to his own credit. These collections are distinct, however, from Parry's 1872 "Alpine Flora of the Rocky Mountains" collection, and from his 1861 collection from "the headwaters of Clear Creek" in Colorado, which foreshadowed his trip with Hall & Harbour. The 1862 Parry specimens are, by all evidence, from the same place as the Hall & Harbour specimens (i.e. Park County and its environs), and the database records were adjusted accordingly.

An additional and related collection, entitled "American Plains Flora Lat 41", was not included in this this locality improvement effort. However, these resources do improve our knowledge of the collecting localities on the route that Hall & Harbour took to and from Colorado, from Omaha to the Colorado border along the Platte River, and thence along the South Platte fork to Denver. Unfortunately, this linear pattern across state lines does not offer much opportunity for improved georeferencing.

Hall & Harbour Update Example

Specimen: PBRU00029889

Species: Aquilegia vulgaris var. brevistyla

Collector Number on Sheet: 23

Writing on Sheet:Aquilegia vulgaris var. brevistyla / Rocky Mountain Flora, Lat. 39°-41° / no. 23 / E. Hall & J. P. Harbour, Colls. 1862”

Current Database Entry: Park County, CO / 1862 / No. 23 / E. Hall + J. P. Harbour / Rocky Mountain Flora, Lat. 39°-41°. From Pike’s Peak in the South to the headwaters of Clear Creek in the North [Park, Jefferson, Clear Creek, Grand, or El Paso County].

Conclusion

By cross-referencing the information on labels attached to specimens with information contained in historical journals and secondary literature, we have drastically improved the precision (and reduced georeferencing area) of our database records. A systematic effort to continue this process would benefit from cooperation between herbaria, especially when no single herbarium has a complete set of specimens from a given expedition. The sources identified here can be used directly and indirectly by other herbaria interested in improving locality data for 19th century western US plant specimens. 

Works Cited

Caffey, David L. (2014) "Chasing the Santa Fe Ring: Power and Privilege in Territorial New Mexico". Albuquerque, NM: University of New Mexico Press.

Despain, Don G. (2007) "A Region of Astonishing Beauty: The Botanical Exploration of the Rocky Mountains by Roger L. Williams", Western North American Naturalist: Vol. 67 : No. 1 , Article 24. Retrieved from https://scholarsarchive.byu.edu/wnan/vol67/iss1/24

Engelmann, George. (1868) "Altitude of Pike’s Peak and other points in Colorado Territory" Academy of Science of St. Louis. v.2 1861-1868. Transactions of the Academy of Science of Saint Louis. [St. Louis: Academy of Science of St. Louis]. 126-133. Retrieved from https://catalog.hathitrust.org/Record/003927962

Gray, Asa (1863). “Enumeration of the Species of Plants Collected by Dr. C. C. Parry, and Messrs. Elihu Hall and J. P. Harbour, during the Summer and Autumn of 1862, on and near the Rocky Mountains, in Colorado Territory, lat. 39°-41°.” Proceedings of the Academy of Natural Sciences of Philadelphia, 15, 55-80. Retrieved from http://www.jstor.org/stable/4059746

Gray, Asa (1849). “Plantae Fendlerianae Novi-Mexicanae: An Account of a Collection of Plants Made Chiefly in the Vicinity of Santa Fé, New Mexico, by Augustus Fendler; with Descriptions of the New Species, Critical Remarks, and Characters of Other Undescribed or Little Known Plants from Surrounding Regions.” Memoirs of the American Academy of Arts and Sciences (New Series) 4(1): 1-116. Retrieved from https://archive.org/details/mobot31753002327622

Jones, William. Stanhope E. Blunt, Theodore B. Comstock, Charles L. Heizmann, Charles C. Parry, and Joseph D. Putnam (1875). "Report upon the reconnaissance of northwestern Wyoming, including Yellowstone national park, made in the summer of 1873." Washington, D.C.: Government Printing Office. Retrieved from https://archive.org/details/reportuponrecon00putngoog

King, Clarence, Sereno Watson, and Daniel C. Eaton (1871). "Report of the Geological Exploration of the Fortieth Parallel, Volume V: Botany." Professional Papers of the Engineer Department, U.S. Army, No. 18. Washington, D.C.: Government Printing Office. Archive of the US Geological Survey. Retrieved from https://pubs.er.usgs.gov/publication/70039231

Parry, Charles C. (1868) “Ascent of Pike’s Peak, July 1st, 1862, by Dr. C. C. Parry. From a Letter addressed to Prof. Torrey, and communicated by him.” Academy of Science of St. Louis. v.2 1861-1868. Transactions of the Academy of Science of Saint Louis. [St. Louis: Academy of Science of St. Louis]. 120-125. Retrieved from https://catalog.hathitrust.org/Record/003927962

Parry, Charles C. (1874). "Botanical observations in Western Wyoming: with notices of rare plants and descriptions of new species collected on the route of the North-Western Wyoming expedition under Captain W.A." Jones. Salem, MA: Salem Press. Hathi Trust Digital Library. Retrieved from https://catalog.hathitrust.org/Record/101745493

Shaw, E. (1982). “Augustus Fendler's Collection List: New Mexico, 1846-1847.” Contributions from the Gray Herbarium of Harvard University (212), 1-70. Retrieved from http://www.jstor.org/stable/41764742

Standley, Paul C. (1910). "The Type Localities of Plants First Described From New Mexico," In Contributions from the United States National Herbarium,13(6). Smithsonian Institution. Washington, D.C.: Government Printing Office. From the collections of University of Michigan. The Internet Archive. Retrieved from https://repository.si.edu/bitstream/handle/10088/27202/usnh_0013.06.pdf

Weber, William A. “Additions to the flora of Colorado - III” (1961). Series in Biology. 19. Retrieved from https://scholar.colorado.edu/sbio/19/