This article was downloaded by: [The Aga Khan University]On: 25 November 2014, At: 07:04Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Ichnos: An International Journal for Plant andAnimal TracesPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/gich20

Corimbatichnus fernandezi: A cluster of fossil beecells from the late cretaceous‐early tertiary ofUruguayJorge F. Genise a & Mariano Verde ba CONICET Laboratorio de Icnología , CRILAR (5301) Anillaco, La Rioja, Argentinab Departamento de Paleontología, Facultad de Ciencias , Universidad de la República ,Iguá 4225, C.P., 11400, Montevideo, UruguayPublished online: 17 Dec 2008.

To cite this article: Jorge F. Genise & Mariano Verde (2000) Corimbatichnus fernandezi: A cluster of fossil bee cells fromthe late cretaceous‐early tertiary of Uruguay, Ichnos: An International Journal for Plant and Animal Traces, 7:2, 115-125,DOI: 10.1080/10420940009380150

To link to this article: http://dx.doi.org/10.1080/10420940009380150

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purposeof the Content. Any opinions and views expressed in this publication are the opinions and views of theauthors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should notbe relied upon and should be independently verified with primary sources of information. Taylor and Francisshall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, andother liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relationto or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

khnos, 2000, Vol. 7(2), pp, 115-125Reprints available directly from the publisherPhotocopying permitted by license only

© 2000 OPA (Overseas Publishers Association) N.V.Published by license under

the Harwood Academic Publishers imprint,part of the Gordon and Breach Publishing Group.

Printed in Malaysia

Corimbatichnus fernandezi: A Cluster of Fossil Bee Cellsfrom the Late Cretaceous-Early Tertiary of UruguayJORGE F. GENISEa* and MARIANO VERDEb

"CONICET Laboratorio de Icnología, CRILAR (5301) Anillaco, ha Rioja, Argentina and ^Departamento de Paleontología,Facultad de Ciencias, Universidad de la República, Iguá 4225, C.P. 11400 Montevideo, Uruguay

Corimbatichnus fernandezi n.igen, n. isp. is a cluster offossil bee cells from the Uruguayan Late Creta-ceous-early Tertiary Asencio Formation. It is com-posed of rows of small excavated cells contained inpaleosol peds. Cells have inner surfaces somewhatrough and are closed with loose material. Separationamong cells is very thin, and the cells are orientateduniformly, showing a great economy of space as insome halictine constructions. Corimbatichnus can bedistinguished from Rosellichnus and Uruguay by itsconvex face, excavated cells and the spatial arrange-ment of them.

Keywords: Bee cells, Halictinae, Asencio Formation, Late Cre-taceous-early Tertiary, Uruguay

INTRODUCTION

Recently, Genise and Bown (1996) described theonly two known types of clusters of fossil beecells, one being, Rosellichnus from the Tertiary ofArgentina and the Arab Emirates, and the other,Uruguay, described originally by Roselli (1938)from the Late Cretaceous-early Tertiary AsencioFormation of Uruguay. In this paper, wedescribe a third ichnogenus, ocurring in theAsencio Formation, where we have undertakenextensive research over the last several years toreview Roselli's work and to study new fossilinsect nests. The new trace fossil is very differentmorphologically from those described previ-

* Correspondence Author.

ously, particularly with respect to the arrange-ment of cells, which show a high economy ofspace as in some halictid "combs" (sensu Sakag-ami and Michener, 1962).

GEOLOGIC AND STRATIGRAPHICSETTING

The trace fossils are present within the extendedLate Cretaceous-early Tertiary outcrops fromwestern Uruguay in southern South America.The geographic location of the quarries can beconsulted in previous papers of this series (Gen-ise and Bown, 1996; Genise and Laza, 1998) Theage, stratigraphic position and other aspects ofthe outcrops that bear the fossil nests is still mat-ter of discussion (Genise and Bown, 1996; Vero-slavsky and Martinez, 1996; Gonzalez et al.,1998; Pazos et al., 1998). The trace fossils occur intiered paleosols developed in the Palacio Mem-ber of the Asencio Formation, about 20 m of fer-ruginous quartz sandstones that occur betweenthe subjacent Cretaceous Mercedes Formationand the Oligocene Fray Bentos Formation. Theage of this member is still controversial becauseof the complete absence of plant or animalremains.

115

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

116 JORGE F. GENISE and MARIANO VERDE

SYSTEMATIC ICHNOLOGY

Corimbatichnus n. igen

Etymology

Derived from the Latin Corymbiatus, meaningclustered, and from the Greek lchnos meaningtrace.

Type ichnospecies

Corimbatichnus fernandezi, type and only knownichnospecies

Diagnosis

Cluster of elongated cells, having rounded rearsand flat tops, excavated in paleosol peds. Cellsare devoid of discrete walls, they have inner sur-faces somewhat rough and loosely packed clo-sures. Cell entrances open in a convex face of theped and they are arranged in rows and columns.Cells in each row are orientated with their axesparallel and inclined with respect to the convexface. The arrangement of cells result in thin inter-cellular separations. Some peds may contain twodifferent clusters, in which case the cells of oneare orientated to form a "V" with respect to theother. Clusters are vertically orientated, havinghorizontal and inclined cells.

Remarks

Corimbatichnus can be distinguished from Uru-guay and Rosellichnus by the absence of con-structed walls, the arrangement of cells and theconvex aspect of the face containing the cellentrances.

Corimbatichnus fernandezi in. isp.

Etymology

Dedicated to Wilfredo Fernandez, curator of thecollection of fossil insect nests housed at theMuseo Lucas Roselli from Nueva Palmira, field

assistant of the late Lucas Roselli and an invalua-ble collaborator in our recent field trips.

Holotype

Museo "Francisco Lucas Roselli" (NuevaPalmira, Uruguay) N° 727 (Maldonado Quarry,Genise-Bown col., 2-1994)

Hypodigm

Paratypes: Museo "Francisco Lucas Roselli"(MLR, Nueva Palmira, Uruguay) N° 732 (EspigaQuarry, Genise col, 5-1994) and N° 735 (EspigaQuarry, Verde col., 10-1995) and Museo Argen-tino de Ciencias Naturales Laboratorio deIcnologia, (MACN-LL Buenos Aires, Argentina)N° 611, 616, 618 and 619 (Maldonado Quarry,Genise col., 2-1995, 10-1996 and 9-1997). Othermaterial examined: fourteen small fragments ofcombs (MLR N° 734 and MACN-LI N° 612, 613,614,615,617,620/628).

Type locality and Occurence

Palacio Member of the Asencio Formation (LateCretaceous-early Tertiary). Maldonado andEspiga quarries near the town of Nueva Palmiraand Estancia Media Agua near the town ofPalmitas.

Diagnosis

Only known ichnospecies, same as for the ichno-genus.

Description

The clusters of excavated cells occur in paleosolpeds, which usually contain a more or less com-plete cluster plus several cells from another.Cells show a partially smoothed inner surface, inwhich it is possible to recognize coarse grains ofthe soil material protruding inside the chamber(Fig. 2 D). The cells show rounded cross-sec-tions, without a distinct flat floor, indicating thatthey have radial symmetry. The rear ends areround and the cells show a constriction or necknear the top, which in some cases is filled with

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

FOSSIL BEE CELLS: URUGUAY 117

FIGURE 1 Corimbatichnusfernandezi. (A) Holotype, front. (B) Holotype, back. (C) Paratype (MLR N° 732) showing the arrangementin "V" of the two combs in the same ped. (D) Paratype (MLR N° 732) front view of one comb. (E) Paratype (MACN-LI N° 616)showing the arrangement of the combs in the ped. (F) Paratype (MACN-LI N° 616) front view of one comb. Scale bar: 1 cm

soil material. No spirally designed cap can be material filling the neck was the original closurerecognized in any cell, suggesting that the soil of the cell. Cells do not show discrete walls. Only

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

118 JORGE F. GENISE and MARIANO VERDE

one incomplete cell of a fragment, at its posteriorend, shows a distinct wall 0.5 mm thick slightlyseparated from the chamber wall (Fig. 2 C). Thiswall is constructed from a finer material but it is

impossible to ascertain if it represents a distinctlining produced by the original constructor ofthe trace or a superimposed structure of anotherspecies of parasite or inquiline.

FIGURE 2 Corimbatichnusfernandezi. (A) Paratype (MLR N° 735) front view. (B) Paratype (MLR N° 735) back view. (C) Fragmentshowing an incomplete cell with a lining. (D) Fragment showing the inner surface of cells. (E), (F) and (G) Holotype in whichsticks show the inclination of cells in an arched row, a straight row and transverse cross-section respectively

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

FOSSIL BEE CELLS: URUGUAY 119

Peds containing the most complete clustershave a more or less flattened cubic design. Meas-urements range from 34 mm to 55 mm in length,from 26.8 mm to 34 mm in width and from20 mm to 33 mm in height. In the best preservedspecimens the cell length ranges from 14 mm to19 mm. (mean: 17 mm, N=24); the cell maximumdiameter ranges from 3.5 mm to 4.7 mm (mean:4.09 mm, N=29); and the cell entrance diameterranges from 2.6 mm to 3.2 mm (mean: 2.93 mm.,N=20). In the best preserved samples, thenumber of cells ranges from 22 to 44 and may bedisposed in one or two groups. To thoroughlyreconstruct the entire morphology of the traceand to interpret the trace maker behavior it isnecessary to integrate the descriptions of severalof the most complete specimens found. No sin-gle example displays the complete morphologyof a mature cluster.

The holotype (Fig. 1 A, B) shows a convex sur-face where the cell entrances can be located in animaginary grid composed of rows {x axis) andcolumns {y axis). It is possible to distinguish 8rows of cell entrances located in concave chan-nels surrounded by two ridges. The longest rowbears 7 cells. In each row (x axis) the cells are ori-entated uniformly, with their longitudinal axesparallel to each other and all of them at an angleof 110° with respect to the tangent of the archformed by the convex face of the ped (Figs. 2 F, 4B). It is also possible to observe that in the long-est row, the longitudinal axis of the first cell isrotated about 30° with respect to the last one(seventh) in a hypothetical z axis (Figs. 2 E, 4 A).Columns of cells are more difficult to define;they can be recognized at two angles (60° or120°) with respect to the rows (x axis). The angleresults from the displacement of the rows insuch a way that the cells of one row are exca-vated over the intercellular space of the neigh-boring rows. This comb-like arrangement is notperfect; the cells do not lay just in the middle ofthe intercellular space of the neighboring rows.In such inclined columns, the cells are equallyorientated with their longitudinal axes parallel

to each other (Fig. 4 D). But also, it is difficult tounequivocally follow a column because of thecurvature of the ped and imperfections of thedesign. When sectioned specimens in a y-z planeperpendicular to the rows of cells, indicate thatthe longitudinal axes are convergent down-wards (Figs. 2 G, 4 C). Probably, the trace is con-structed by the sucessive addition of rows andthus these can be better defined than the col-umns, which are not the direct result of thebuilding behavior but a secondary consequenceof it. The channels which define the rows appearto be the remains of the tunnels where the cellswere excavated from. The missing half of thetunnel would have been in the surroundingpaleosol (cf. Ellipsoideichnus; Genise and Hazel-dine, 1998a). The back part of this cluster showsthe remains of at least four cells clearly belong-ing to another one excavated at a very differentangle to those described above. The longitudinalaxes of these cells are almost parallel to the y axisof the grid and so at an almost right angle withthe longitudinal axes of the cells of the formergroup.

Specimen N° 616 (MACN-LI) (Fig. 1 E, F)shows the arrangement of two groups of cells inthe same ped. The best preserved group is com-posed of five incomplete rows of cells, whichdisplay the same disposition described previ-ously (Fig. 1 F). The second group is composedof the remains of four rows of incomplete cells.The broken extremes of the ped show the cells ofboth groups orientated with their longitudinalaxes forming a "V" (Fig. 1 E). At one extreme,the axes are orientated almost at a right angleand the rear ends of the cells of both groups areseparated by thin spaces, whereas at the other,angles may reach 140° and some interspaces arelarger. Specimen N° 732 (MLR) also showsclearly the presence of two groups of cells in thesame ped at an angle of 140°-150° (Fig. 1 C, D).The ped in the upper part of the "V" is broken,but it is possible to recognize three very incom-plete cells orientated upwards, which are notclearly related to any of the two groups. One

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

120 JORGE F. GENISE and MARIANO VERDE

possibility is that both groups of cells were sepa-rated by a hiatus; the other possibility is thatboth were connected by additional rows of cells.In a way, the broken extremes of a ped show asection in the y-z plane perpendicular to therows, and even in the same group of cells thelongitudinal axes show a convergent array asdescribed previously.

Specimen Nc 735 (MLR) comprises 6 rows ofcells repeating the same arrangement (Fig. 2 A,B). A second group of cells is indicated only bythe remains of one cell orientated at a right anglewith respect to the group. In this specimen agroove is preserved in the back part (Fig. 2 B),and this may represent the trace of a root wherethe ped would be attached like other fossil nestsof this formation (cf. Uruguay, Genise and Bown,1996; Monesichnus, Genise and Laza, 1998).

The other specimens examined show the samemorphology or are very fragmentary. Most werefound as isolated clasts, broken or weathered, onthe floor of the quarries. In most cases it isimpossible to distinguish unequivocally if thesespecimens with few cells represent youngeststages of combs, which would permit study ofthe development of the whole structure. Othersoccur.in individual peds of the blocky subhori-zon of the paleosol. In many cases these pedsmay be rotated and weathered (Gonzalez et al,1998) precluding any conclusion with respect tothe original position of the trace in the soil aswell,as its complete morphology.

DISCUSSION AND CONCLUSIONS

Coritnbatichnus fernandezi poses several problemsfor discussion; among others, the reconstructionof the architecture of the whole trace, itscomb-like architecture, the original position inthe paleosol and the identity of the trace maker.

All the specimens examined show the samearrangement of cells. Two groups excavated inthe same ped are orientated in the form of a "V".This orientation leaves no doubt about the origi-

FIGURE 3 3D reconstruction of C. fernandezi showing (A) thegeneral aspect and (B) a cross-section of the ped showing thearrangement of the cells

nal position of the ped and the cells in the soil.Bee cells may be vertical, inclined or horizontalbut usually the posterior ends are orientateddownwards to avoid the spilling of provisions.Hence, the only possible position of C. fernandezi

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

FOSSIL BEE CELLS: URUGUAY 121

FIGURE 4 3D reconstruction of C. fernandezi. (A) Front view of a comb in which some cells were removed to show particularfeatures. Note the rotation of the longitudinal axes of the cells in the row below. (B) Cells inclined uniformly in a row. (C) Diver-gence of cells in a cross-section of the comb perpendicular to the rows. (D) Cells inclined uniformly in a column (120° from arow)

in which no cell would be orientated bottomupwards is that of a "V", being each group ofcells vertically orientated on both sides of theped (Fig. 3 A, B). In each group, cells are orien-tated uniformly and they are separated by thinwalls; the two requisites that Sakagami andMichener (1962) proposed to consider a group ofcells of a halictine as a true "comb". For theseauthors, the combs represented the most effi-

cient way to achieve space economy and theycould be excavated in a ped or constructedinside a cavity previously performed by the bee.In more Tecent papers on halictine architecture(Eickwort and Sakagami, 1979; Engel, 1998),combs are no longer distinguished from otherclusters. It is impossible to ascertain to date ifboth clusters of a ped were originally connectedby additional rows of cells, in which case the

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

122 JORGE F. GENISE and MARIANO VERDE

complete morphology of C. fernandezi would bethat of a single cluster excavated in a ped.

The convex surface of the clusters indicatesthat they were surrounded by a cavity con-structed previously by the bee. Sakagami andMichener (1962) compared different nest pat-terns in halictinae from the standpoint of spaceeconomy. Some of them may be illustrated usingthe clusters of fossil bee cells described from thesame Asencio Formation (Fig. 5). In the firstinstars of the series, the cells open directly intothe main burrow, resulting in clusters havingcells with close entrances and divergent bottoms(Fig. 5 A). In these clusters the face where thecells open is concave because it reflects the out-line of the tunnel wall. In addition, there is littlespace economy because of the misuse of inter-spaces among the bottoms of the cells. Such isthe case of the ichnogenus Uruguay. A secondstage in this series is exemplified by Rosellichnus:the frontal face of the cluster is flat, reflecting theexistence of a previous cavity from where toexcavate the cells (Fig. 5 B). Nevertheless, inRosellichnus the interspaces among cells is stillconsiderable. Corimbatichnus shows a convexfrontal surface, revealing that a cavity was alsopreviously excavated, but additionally it showsthe most efficient inclination of cells in terms ofspace economy. When the cells are excavatedfrom a convex surface, large spaces among cellsare necessary to avoid bottom overlapping, if thelongitudinal axes are coincident with the radiusof the curvature (Fig. 5 C). Economy of space isonly achieved if the cells are excavated at a par-ticular angle with respect to this curvature, as inCorimbatichnus (Fig. 5 D). Although efficient, thisarchitecture is limited in terms of cluster expan-sion: the addition of more cells in the row is pos-sible as far as cell bottoms begin to overlap(Fig. 5 E). To avoid this overlap a possible solu-tion is to make use of the 3D space by coiling thecell row to form a helix, as in Ellipsoideichnus(Fig.5F).

The identity of the trace maker is in littledoubt and, in addition, contributes to the pale-

ontological record of the family. Clusters sur-rounded by a cavity, with few exceptions, areperformed by Halictinae, the sweat bees (Sakag-ami and Michener, 1962). Other features ofCorimbatichnus fernandezi also confirm these con-structors. A distinctly constricted neck as in cellsof C. fernandezi is characteristic of cells of sweatbees. In contrast, the bilateral symmetry of halic-tinae cells is not present in cells of C. fernandezi,which seem to be radially symmetrical. Theabsence of a spiral closure, typical of most beecells, is also common in halictinae cells. Thesebees close the cells with a permeable, loose, tam-pon-like closure according to Sakagami andMichener (1962), similar to that found in C. fern-andezi. Another exception to a common bee cellfeature, the absence of a smooth inner lining, canbe explained in terms of halictinae architecture.With some exceptions, most sweat bees excavatecells in sandy soils and then line them with dif-ferent materials: clay or fine wood particles, plusa secreted waxlike substance. The rough innersurface of C. fernandezi may indicate that thesmooth inner lining had yet not been performed,that it was not preserved or that, as in some Aus-tralian species that nest in sandy soils, such lin-ings were not constructed at all (Sakagami andMichener, 1962). Because in most bees thesmoothing of the inner surface of the cells issimultaneous with its construction, the absenceof this smoothed surface in cells of C. fernandezimay be other evidence of the halictinae origin ofthis trace fossil.

Engel (1996) found that the known older halic-tids came from Eocene-Oligocene deposits, butbased on phylogenetic evidence, this author,estimated that the augochlorines should be 75-80 millions years old. Later, Engel (1998) pro-posed that these primitive augochlorines wouldhave been semisocial species, constructing cellclusters in soils, and mentioned as physical evi-dence of this behavior the ichnogenus Uruguay.However, as discussed in previous papers (Gen-ise and Bown, 1996; Genise and Hazeldine,1998b), in spite of being clusters of cells, Uruguay

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

FOSSIL BEE CELLS: URUGUAY 123

FIGURE 5 Different levels of space economy in clustered cells. (A) Black triangles show misused spaces between convergentcells (concave clusters). (B) The expansion of the tunnel results in a cluster having a flat face and more space economy. (C) Blacktriangles show misused spaces between divergent cells (convex clusters). (D) Cells inclined uniformly in a convex cluster avoidthe space misuse. (E) In convex clusters the addition of cells in a row is limited by overlapping (black area) an the exhaustion ofthe free space (circle). (F) The helicoidal rows of cells avoid their overlapping

cannot be attributed to sweat bees because of itssize, thickened walls, spiral closures and otherfeatures. In contrast, the size, thin walls andother features of C. fernandezi closely resemble

those of halictinae clusters, and would representdefinitively the pre-Oligocene evidence of theseconstructions.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

124 JORGE R GENISE and MARIANO VERDE

Among halictines, constructions such as C.fernandezi are not restricted to a particular taxon(Sakagami and Michener, 1962; Eickwort andSakagami, 1979). They are known in species ofunrelated genera belonging to Augochlorini andHalictini, albeit Eickwort and Sakagami (1979)stated that clustering of cells is predominant inthe neotropical Augochlorini and relativelyuncommon in the cosmopolitan Halictini. Theclusters of some of them are different fromCorimbatichnus in having the cells vertically ori-entated (Augochloropsis, Megommation), divergent(Paroxystoglossa, and some Caenaugochlom andLasioglossum), or in having flat or concave faces(some Lasioglossum, Augochlorodes, Augochlorella).According to the data compiled by theseauthors, vertical clusters are constructed by spe-cies of Augochlorella, Halictillus, Corynura s. str.,Paroxystoglossa, Augochlorodes and Pereirapis.Another feature shown by Corimbatichnus, theexistence of two opposite clusters in the sameped, is recorded exceptionally in the SouthAmerican Corynura cristata and Paroxystoglossajocasta (Sakagami and Michener, 1962). In sum-mary, Corimbatichnus combines features exhib-ited by different clusters belonging to unrelatedspecies of Halictinae, but without fitting in anyof them in particular.

In this context it is difficult to speculate aboutthe sociability of the constructor. Halictinaeinclude solitary to primitively eusocial species,even belonging to the same genus (Michener,1974). Eickwort and Sakagami (1979) concludedthat among Augochlorini, the typical clus-ter-making halictines, semisocial behavior iscommon, whereas in Halictini larger eusocialsocieties are widespread. Later, Engel (1998)found that the semisocial condition would be aprimitive condition for augochlorines. However,the fact is that simple nests are constructed bysolitary species as well as social ones and, simi-larly, clusters belong indistinctly to solitary andsocial species (Sakagami and Michener, 1962).

C. fernandezi was probably constructed by asingle species of bee as suggested by the similar

size of all the complete cells measured. Thus, thisrecord adds another species of bee and anothertrace fossil to those previously documented fromthe Asencio Formation. These trace fossil associ-ations are composed mainly of dung-beetle andbee nests, which are indicative of open herba-ceous environments (Genise and Bown, 1994).Genise and Bown (1996) originally believed thatthe paleosols that bear the trace fossils in theAsencio Formation were oxisols that probablyhad developed under humid conditions, similarto those of tropical rain forests. However, recentstudies (Gonzalez et al., 1998) postulated that thestructure of these paleosols more likely reflectssoils of open environments, in agreement withthe abundance of dung-beetle and bee nests.

Acknowledgements

We thank Gerardo Genuario, Wilfredo Fernan-dez and Jorge Frugoni from the Museo LucasRoselli for their assistance during different fieldtrips since 1994, when the first specimen of thistrace fossil was collected. Also Jerome G. Rozenand Michael S. Engel (AMNH) for the criticalreview of the manuscript and the Ichnos review-ers Andre Nel and Christian Gaillard. Thisresearch was partially supported by a grant ofFundacidn Antorchas to the senior author.

References

Eickwort, G.C. and Sakagami, S.F. 1979. A classification ofnest architecture of bees in the tribe Augochlorini(Hymenoptera: Halictidae: Halictinae) with descriptionof a Brazilian nest of Rhinocorynura inflaticeps. Biotropica,11: 28-37.

Engel, M.S. 1996. New augochlorine bees (Hymenoptera:Halictidae) in Dominican amber, with a brief review offossil halictidae. Journal of the Kansas EntomologicalSociety, 69: 334-345.

Engel, M.S. 1998. Phylogeny, classification and evolutionaryethology of the bee tribe Augochlorini (Hymenoptera,Halictidae). Ph.D. Dissertation, Cornell University, Ith-aca, New York, USA. xxii+306 p.

Genise, J.F. and Bown, T.M. 1994. New Miocene scarabeidand hymenopterous nests and Early Miocene (santacru-cian) paleoenvironments, Patagonian Argentina. Ichnos3: 107-117.

Genise, J.F. and Bown, T.M. 1996. Uruguay Roselli 1938 andRosellichnus, n. ichnogenus: two ichnogenera for clusterof fossil bee cells. Ichnos, 4: 199-217.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014

FOSSIL BEE CELLS: URUGUAY 125

Genise, J.F. and Hazeldine, P.L. 1998a. 3D-reconstruction ofinsect trace fossils: Ellipsoidekhnus meyeri Roselli. Ichnos,5: 167-175.

Genise, J.F. and Hazeldine, P.L. 1998b. The ichnogenusPalmiraichnus Roselli for fossil bee cells. Ichnos, 6: 151-166.

Genise, J.F. and Laza, J.H. 1998. Monesichnus ameghinoiRoselli: a complex insect trace fossil produced by twodistinct trace makers. Ichnos, 5: 213-223.

González, M, Tófalo, O.R. and Pazos, P. 1998. Icnologia ypaleosuelos del Miembro del Palacio de la FormationAsencio (Cretacico Superior-Terciario Inferior) del Uru-guay. Actas del II Congreso Uruguayo de Geología. 38-42.

Michener, C.D. 1974. The social behavior of bees. TheBelknap Press of Harvard University, Cambridge, Mas-sachusetts, 404 p.

Pazos, P., Tófalo, O.R. and González, M. 1998. La paleosuper-ficie Yapeyú: significado estratigráfíco y paleoambientalen la evolución del Cretácico Superior del Uruguay.Actas del II Congreso Uruguayo de Geología: 59-63.

Roselli, F.L. 1938. Apuntes de geología y paleontología uru-guaya. Sobre insectos del Cretácico del Uruguay odescubrimiento de admirables instintos constructivos deesa época. Boletín de la Sociedad Amigos de las CienciasNaturales "Kraglievich-Fontana" 1: 72-102.

Sakagami, S.F. and Michener, C.D. 1962. The nest architec-ture of the sweat bees (Halictinae). The University ofKansas Press, 135 p.

Veroslavsky, G. and Martínez, S. 1996. Registros no deposita-cionales del Paleoceno-Eoceno del Uruguay: nuevoenfoque para viejos problemas. Revista de la Universi-dade Guarulhos, (Serie Geociencias), 1: 32-41.

Dow

nloa

ded

by [

The

Aga

Kha

n U

nive

rsity

] at

07:

04 2

5 N

ovem

ber

2014