Sam (and others),
I have some comments inspired by your recent posting of Nomada updates and
Your mention of "Database records available on the website Discoverlife
(www.discoverlife.org)" for N. fervida is perfectly acceptable, but some
additional clarification may help dispel ongoing, widespread
misunderstanding about the status of "Discover Life" data. I often receive
feedback suggesting a widespread and persistent belief that all bee
records mapping on Discover Life maps are owned or validated by "Discover
Life" or even by me personally. This is certainly not the case. It is very
important to consider carefully the various data sources and the merits
and limitations of each source individually.
All data on Discover Life for N. fervida and all other species have a
particular data owner other than Discover Life. For a large proportion of
records, matters are further complicated because these data have
previously been delivered to and served by an additional data integration
portal (i.e. GBIF). Thus further clarification is needed of the
relationships between the data display portal in question (DL), the
additional data integration portal (GBIF), and primary data owners, and
how these relationships complicate efforts to maintain consistency of
nomenclature and data quality. Anything Sam and others can do to clarify
these relationships when referring to database records available on
www.discoverlife.org will be much appreciated, as this may encourage
people to think about the complexities of data ownership, maintenance,
quality control, citation, etc. Consideration of such complexities is
especially important when taxonomy is in flux as is the case with Nomada
fervida and texana sensu lato.
Of particular relevance to the Nomada update, I hope people will be aware
of these issues when viewing maps of Nomada texana after I update the few
databases I am actually responsible for (the AMNH BEE specimen database,
the AMNH BEES literature database, my image database, and the Bee species
guide) to reflect the forthcoming split of Nomada texana sensu Mitchell
into two species. When this happens, I can immediately update "my"
databases, but unfortunately there is no efficient mechanism for me or
anyone else to simultaneously update records directly in GBIF and/or in
any or all source databases of GBIF providers. Unless great improvements
are made to existing bioinformatics systems, updates will not happen
quickly even under the best case scenario, i.e. if I (or better yet
someone else) were to immediately send feedback to all GBIF data owners
individually and they were to immediately make all needed updates to their
source databases. Even if this happened with unprecedented speed and
accuracy there would still be no way to ensure prompt delivery of the
updated data to GBIF, global updating of the record across GBIF itself,
and timely availability of the updated GBIF data for processing onto
Discover Life maps. I hope everyone will consider such challenges when
they view Discover Life maps and evaluate my responsibility or lack
thereof for perceived shortcomings of these.
Please consider that it is best to consider data providers individually,
most useful to send feedback directly to data providers as necessary using
the link at the bottom of all records, essential to individually credit
source database (not just data integration and display portals such as
Discover Life and GBIF), and optimal to contact these directly whenever
complete and accurate data and metadata are required.
Use of the "Make Map" function in Discover life as opposed to the default
Global Map (optimized for speed of loading) is helpful. Using "Make Map"
it is possible to add or delete data sources as desired so as to view only
the most relevant data in the most attractive way, with greater control
than is possible with the default map. For example, it is useful to delete
the AMNH BEES literature (state-level) database in the many cases where
only precise specimen records are desired. It is often useful to view only
those databases known to have been most recently updated to reflect
current taxonomy or, on the other hand, to view only databases most in
need of updating so that their records are conspicuous.
Regarding synonymy of N. fervida sensu lato, at Cornell I studied at least
one specimen collected together with A. splendens from near Lake Ontario
in upstate NY. I strongly suspect that a dark form of true N. fervida
occurs along the Great Lakes into Ontario, Canada, so am not surprised
that your Ontario specimen matches Florida fervida. However, I am less
confident that such dark "fervida" are conspecific with topotypical N.
wisconsinensis from Wisconsin. Specifically, I am not convinced that the
darker form of N. fervida from upstate NY is conspecific with the N.
wisconsinensis from U. of WI Madison I studied recently (reported by Wolf
and Ascher, 2009). Unfortunately I have never been able to make a direct
comparison of the relevant specimens. Lacking this, I would like to see
DNA evidence from closer to the type locality of wisconsinensis so as to
better prove the synonymy. Ontario is outside the previously known range
of either fervida or wisonsinensis so I am reluctant to accept DNA from an
Ontario specimen as certainly representative of wisonsinensis.
Note that Andreas Mueller's attractive website "Palaearctic Osmiine Bees"
mentioned by sam with refence to synonymy of Osmia hyperborea, is a very
useful and reliable resource for all Osmiini species shared with the Old
World (not just subgenus Melanosmia), namely:
Chelostoma (Foveosmia) campanularum (Kirby, 1802)
Chelostoma (Gyrodromella) rapunculi (Lepeletier, 1841)
Hoplitis (Formicapis) robusta (Nylander, 1848)
Hoplitis (Hoplitis) anthocopoides (Schenck, 1853)
Osmia (Helicosmia) caerulescens (Linnaeus, 1758)
Osmia (Melanosmia) inermis (Zetterstedt, 1838)
Osmia (Melanosmia) laticeps Thomson, 1872
Osmia (Osmia) cornifrons (Radoszkowski, 1887)
Osmia (Osmia) taurus Smith, 1873
The link is:
I also very highly recommend the following non-North American bee links
(among very many others):
[follow "Pollinators" link at lower left]
Many of these sites treat many North American species so are of more than
touristic interest. Please enjoy these outstanding websites and share with
Some of you may be interested to know that we have changed the recommended
citation of the former "Apoidea species guide" that serves as the
authority file for Discover Life species pages, maps, and (usually) the ID
The new citation for this compilation is:
Ascher, J. S. and J. Pickering. 2010.
Bee species guide (Hymenoptera: Apoidea: Anthophila).
Draft-22, 27 January, 2010
We changed "Apoidea" to "bee" because Apoidea includes the apoid wasps
(Sphecidae sensu lato) in addition to bees. Data for North American (sensu
lato) apoid wasps are still avaialable on Discover Life, but are now in a
separate "Apoid_wasp_species" guide, as it was confusing to maintain
global bee data with regional wasp data in one guide.
Finally, I wonder why anyone performs N-J analyses when these are
demonstrably inferior to rapid cladistic methods as shown by the paper
cited below (see P.S.)?
I welcome feedback from anyone on any of the above.
P. S. The N-J inferiority paper link:
And the title and abstract:
PARSIMONY JACKKNIFING OUTPERFORMS NEIGHBOR-JOINING
James S. Farris 1 , Victor A. Albert 2 , Mari Källersjö 1 , Diana Lipscomb
3 , Arnold G. Kluge 4 1 Molekylärsystematiska laboratoriet,
Naturhistoriska riksmuseet, Box 50007, Stockholm, S 104 05, Sweden 2 New
York Botanical Garden, Bronx, New York, 10458-5126, U.S.A. 3 Department
of Biology, George Washington University, Washington, D.C. 20052, U.S.A.
4 Division of Reptiles and Amphibians, Museum of Zoology, The University
of Michigan, Ann Arbor, Michigan, 48109-1079, U.S.A.
Copyright 1996 The Willi Hennig Society
AbstractBecause they are designed to produced just one tree,
neighbor-joining programs can obscure ambiguities in data. Ambiguities can
be uncovered by resampling, but existing neighbor-joining programs may
give misleading bootstrap frequencies because they do not suppress
zero-length branches and/or are sensitive to the order of terminals in the
data. A new procedure, parsimony jackknifing, overcomes these problems
while running hundreds of times faster than existing programs for
neighbor-joining bootstrapping. For analysis of large matrices, parsimony
jackknifing is hundreds of thousands of times faster than extensive
branch-swapping, yet is better able to screen out poorly-supported groups.
> USGS Native Bee Inventory and Monitoring Lab Update
> Osmia hyperborea = O. laticeps
> Molly Rightmyer and Mike Arduser alerted me to the fact that Osmia
> hyperborea has been synonymized and is now O. laticeps (Nilsson 2009).
> This species is a northern boreal species.
> John Ascher provided a link to a nice web site with additional information
> on the species and other Osmia in that group:
> The Discoverlife guides have been updated.
> Andrena robervalensis = A. runcinatae
> John Ascher as well as Larking, Jean, and Arduser mentioned that uncommon
> Andrena robervalensis has been synonymized with A. runcinatae
> (Gusenleitner et al. 2005).
> The Discoverlife Guides have been updated.
> Megachile integrella
> This species was on our working list for species that had not been seen in
> the last 20 years. It is a rare species of central Florida and apparently
> Dune or at least very coastal areas along the Atlantic. I had been in
> correspondence with Mark Deyrup regarding when it was last seen at
> Archbold Biological Station in Florida (1987)and after sending out that
> email I looked at a handful specimens that Edd Barrows had bowl collected
> in that general area last May (2009). In that batch was one nice M.
> ingegrella! from a remnant scrub area near Lake Placid. I think this
> demonstrates that many of these "missing" species are simply missing due
> to lack of collecting and that small efforts like Edd's to collect
> wherever you go can have large returns.
> Bees per Bowl - Boston Harbor Islands
> Jessica Rykken, who is coordinating biodiversity surveys on the islands in
> the Boston Harbor of Massachusetts (and a couple of the peninsulas)sent in
> a summary of the 94 bowl transects that they ran over the past several
> years. If you assume they had 15 bowls per transect (in reality it was
> usually one or two less due to loss of bowls or disturbance) they had 2.67
> bees per bowl. However, they had one huge outlier with a transect in
> October getting 732 bees (mostly female Ceratina), if you eliminate that
> transect the average goes down to 2.2 bees per bowl.
> Nomada Synonymies
> Molly Rightmyer, Sean Brady, Cory Sheffield, and myself have been plugging
> away on a set of Nomada synonymies and are just about ready with our
> manuscript. We have reported on some of the species in the past, but
> below are reports for how we are handling the groups N. fervida, N.
> wisconsinensis, N. texana, N. heiligbrodtii (note that we are resurrecting
> the name N. tiftonensis and that the N. texana group is complicated due a
> mixed up history of synonymies. Comments are always welcome.
> Nomada fervida
> Diagnosis.?Of all the Nomada species in eastern North America with a
> distinct posterior pointing spine on the front coxa, N. fervida is the
> only one whose females have such a high number and density of hairs lining
> the outer apical margin of the hind tibia. Unlike the spaced hairs of
> other species that form a single line along the rim, these hairs number
> over 20 and form a tightly packed group without any spaces, looking a bit
> like a bundle of tiny pencils with slightly reddish erasers at the top.
> Also helpful in identifying this species is the yellow integument on the
> scutellum and the entirely black propodeum and propodeal triangle.
> Males of N. fervida also are identified by the combination of the spine on
> the front coxa and extremely dense, numerous, stout, reddish hairs on the
> hind tibia. However, following the pattern of most Nomada species, the
> hairs on the hind tibia of the males are not as prominent and consequently
> must be inspected closely.
> Molecular results.?We obtained DNA barcoding data for five specimens, two
> males and two females from Florida and one male from Ontario. The male
> from Ontario matches the description of N. wisconsinensis, having more
> extensive yellow maculations on the metasomal terga. All of the specimens
> from Florida have maculations that are tinged with orange-brown rather
> than yellow.
> Variation.?Specimens of N. fervida vary primarily in the color and pattern
> of their maculations. The variability can be roughly characterized
> according to geographic distribution, and there appears to be two main
> areas where the species has been collected, one centered in the Great
> Lakes region and one centered in Florida and Georgia. In keeping with the
> general trend of many bees and wasps from the Deep South, the southern
> specimens of N. fervida all have burnt orange overtones to the yellow
> maculations, while in the northern specimens those overtones are much less
> prominent (although still present to varying degrees). On average,
> southern N. fervida tend to have less extensive maculations on the head,
> mespisternum, and metasomal terga and sterna; however, there is extensive
> overlap and similar patterns can be found in both populations. When
> present, the placement of these maculations is identical in specimens of N
> . fervida from both geographic locations and there appears to be no
> significant intraspecific variation in the punctation, body size, relative
> lengths of the flagellar segments, or number of specialized hairs on the
> outer apical margin of the hind tibia of this distinctive species.
> Distribution.?Nomada fervida is an uncommon sand specialist with two
> separate populations, one centered on the Great Lakes and the other from
> the sand areas of Florida up just to the north in Georgia. Database
> records available on the website Discoverlife (www.discoverlife.org)
> indicate that the species has been collected outside of these main areas,
> but we have not confirmed any records aside from the ones listed below.
> Interestingly, all of the southern collection records for N. fervida only
> appear to come from the interior of Florida and none from equally sandy
> coastal dune systems (with the only possible exception of a single record
> from St. Petersburg, Florida). As with all uncommonly collected Nomada,
> we wouldn't be surprised if more focused collecting efforts result in
> additional specimens from similar areas of deep sand, for example in the
> Sandhills of the Carolinas, New Jersey Pine Barrens, or comparable areas
> in Long Island and Cape Cod. However, it is interesting to note that T.B.
> Mitchell (Mitchell 1960, 1962) did not collect this species in North
> Carolina despite having grown up collecting amidst the large deep sand
> deposits in the sandhills and regularly visiting the coastal areas of that
> state. Similarly, we find it strange that this species appears to be
> absent from the extensive Nebraska Sand Hills region as well as other
> interior dunes on the western side of the Great Plains.
> Comments.?We synonymize N. wisconsinensis (representing the northern
> distribution of the species) with N. fervida (representing the southern
> distribution) based on both morphological and DNA barcoding evidence. An
> examination of the two female primary types showed that diagnostic
> characters, especially those of the mandibles, labrum, propodeum, hind
> tibia, and metasoma, are the same between the two specimens (the N.
> wisconsinensis lectotype lacks antennae). The two type specimens appear
> only to differ in the strength and pattern of yellow maculations, with the
> N. wisconsinensis lectotype a brighter yellow (the N. fervida holotype
> appears to be faded, possibly due to preservation), and with the N.
> fervida holotype having more restricted maculations on T3-T5. Nonetheless,
> the maculation patterns on the metasoma are very similar for both types,
> and are the same on both the head and mesosoma. The color pattern
> differences used to justify the recognition of two distinct species by
> previous authors falls well within known patterns of northern and southern
> populations and intraspecific patterns found in other Nomada. Additional
> support for the synonymy of these two species is provided by the nearly
> identical DNA sequences from specimens conforming to the type specimens
> that were caught at both extremes of the species? distribution (i.e.,
> Ontario and Florida, see above).
> We encourage researchers to investigate the large sand deposits of central
> and eastern North America in June and July, at time when these areas are
> not commonly visited, to look for more specimens of N. fervida. Note that
> populations in Florida are out from March until November. While the host
> for this species is unknown, we suspect that Agapostemon splendens may be
> a good candidate. Similar species of Nomada are known to parasitize other
> Agapostemon species (as well as Nomia and Exomalopsis) and A. splendens is
> similarly restricted to sandy areas.
> Nomada texana
> Diagnosis.?Nomada texana is most similar to N. tiftonensis, but can be
> differentiated from the latter species by the following characters: in
> both males and females, S3 and S4 each has a narrow, ivory or very pale
> yellow, transverse maculation that medially curves slightly anteriorly.
> The maculations are slightly interrupted medially or nearly so, and one
> observed specimen from Texas has maculations restricted to small areas
> laterally on S3 and S4. In contrast, in N. tiftonensis there are no
> maculations on any of the metasomal sterna. Also, in both males and
> females of N. texana on T3 there is an uninterrupted, yellow or ivory,
> transverse maculation, while in N. tiftonensis the maculation on this
> tergum is widely medially interrupted, although there is often a faint
> area of paler brown integument connecting the two lateral yellow
> maculations, and in one specimen from Ontario the maculation is entirely
> uninterrupted. Females of the two species can be differentiated by
> characters of the hind tibia (viewed at high magnification): in N. texana
> , there are three to five, very wide and flat, transparent white to
> brownish hairs (depending on the light reflection) whose apical tips are
> squared-off and extend only to the same length as, or slightly further
> than, the surrounding finer, white hairs; while in N. tiftonensis these
> specialized hairs number from seven to ten, increase more dramatically in
> size posteriorly along apical margin, and are clearly longer than the
> surrounding finer white hairs, comparatively thinner than those of N.
> texana (though still thicker than in many other Nomada species), round in
> cross-section, and translucent yellow throughout most of their length but
> at least a few of the longest and posterior-most hairs have the bases
> translucent white and the apical tips opaque brown to dark red. Females
> of N. texana and N. tiftonensis can additionally be separated by the
> punctation of the clypeus relative to the paraocular area: in N. texana
> the punctures on the clypeus are distinctly smaller and denser than those
> on the paraocular area (punctures ranging in size from relatively small to
> large), while in N. tiftonensis the punctures are almost exactly the same
> size and density on the clypeus as in the yellow portion of the paraocular
> area (punctures uniformly large). Males of the two species are not as
> distinctly separated by the punctures of the clypeus, but in general the
> punctures are denser in N. texana than in N. tiftonensis (with significant
> overlap between the two species).
> Males and females of N. texana and N. tiftonensis generally differ,
> sometimes subtly, in features of the antenna: the flagellar segments of N
> . texana are somewhat shorter than those of N. tiftonensis, with F1
> subequal to the length of F2 in N. texana; the antenna (especially the
> scape, pedicel, and F1) is comparatively brighter yellow-orange in N.
> texana than in N. tiftonensis, with the antenna of most N. tiftonensis
> specimens dark brown with black on the posterior surface and dull orange
> markings on the anterior surfaces of the scape, pedicel and F1 (antenna of
> N. texana with much less contrast between the anterior and posterior
> surfaces). As discussed above, the density of punctures on the clypeus
> tends to be much lower in females of N. tiftonensis than in N. texana
> (length of space between punctures often greater than one puncture
> diameter in N. tiftonensis, spaces between punctures rarely greater than a
> puncture diameter in N. texana), although in both species there is
> sometimes an impunctate longitudinal line at the center of the clypeus
> (this impunctate line sometimes very broad in N. tiftonensis). The
> metasomal maculations of several specimens of N. texana are clearly ivory
> and most are pale yellow to near ivory, while in N. tiftonensis the
> maculations are entirely yellow. All observed specimens of N. tiftonensis
> have extensive amounts of black integument on the hind coxa, while in N.
> texana all but a few lack black integumental color and are instead
> entirely orange except for a yellow maculation present in both species,
> however, in a few specimens of N. texana there is a limited amount of
> black at the very base of the hind coxa.
> Molecular results.?We do not have any representatives of N. texana in the
> N-J analysis. Several specimens were included that had previously been
> identified as N. texana, but upon closer examination they are in fact
> representatives of N. tiftonensis.
> Variation.?Both of these very similar species have relatively low degrees
> of variability in most character states. In addition to the information
> presented in the description section, the extent of yellow/ivory on the
> male supraclypeal area varies to a small extent inN. texana with all male
> specimens having pale maculations on the supraclypeal area, although the
> extent varies to a small degree.
> Distribution.?Nomada texana has a more southern and western distribution
> relative to N. tiftonensis, with specimens seen from Arizona and Idaho
> east to Texas and Indiana; we have also seen specimens from Georgia and
> Material examined.?See spreadsheet, specimens are in the main cabinet in
> the Droege section , TX, ID, NV, AZ, and IN
> Comments.?Mitchell (1962) synonymized N. tiftonensis and N. modesta
> rivertonensis with N. heiligbrodtii, and then separated N. texana from N.
> heligbrodti based on the maculation pattern of T2 and T3. Based on the
> characters presented herein we disagree with his conclusions and instead
> believe that N. tiftonensis and N. modesta rivertonensis represent males
> and females of the same species, and that N. heiligbrodtii is a junior
> synonym of N. texana. Mitchell's (1962) keys to males and females of
> Nomada, as well as previously identified specimens in collections, reflect
> the general confusion between these two very similar species.
> Based on the available material, N. tiftonensis appears to inhabit eastern
> sandy areas; all of the collection locales represented by the material,
> with which we are familiar, meet that general description. We are not
> familiar with the collection locales represented by the examined N. texana
> specimens and are therefore unable to comment upon them. However, as
> there is overlap in the distributions of the two species, it would be
> interesting to investigate the differences in the preferred sites. Hosts
> for both species are unknown.
> Nomada tiftonensis
> Diagnosis.?Males and females of N. tiftonensis are most similar to N.
> texana. See Diagnosis under that species for identifying characters.
> Distribution.?Nomada tiftonensis is known from Georgia north to Maine and
> west to Ontario and Wisconsin.
> Material examined.?See spreadsheet, specimens are in the main cabinet in
> the Droege section. Specimens for N. tiftonensis were collected in ME,
> GA, NJ, NH, IN, MD, ONT, SC, MI, NY, and WI.
> Molecular results.?We have five representatives of N. tiftonensis in the
> N-J analysis from widely separated localities (i.e., Ontario, Maryland,
> and South Carolina). All sequences obtained from these specimens were
> less than XX percent divergent.
> In N. tiftonensis, most male specimens have a completely black
> supraclypeus, others have vague yellowish patches present, and a few have
> solid yellow blocks. In male N. tiftonensis the labrum varies from
> completely yellow to nearly all dark; the labrum in all N. texana males
> was completely yellow. See further comments about observed character
> variation under N. texana.
> Comments.?See Comments under N. texana.
John S. Ascher, Ph.D.
Bee Database Project Manager
Division of Invertebrate Zoology
American Museum of Natural History
Central Park West @ 79th St.
New York, NY 10024-5192
work phone: 212-496-3447
mobile phone: 917-407-0378