SciELO - Scientific Electronic Library Online

vol.35 issue5Nesting biology of Centris flavifrons (Friese) (Hymenoptera: Apidae, Centridini), one of the main pollinators of Byrsonima crassifolia L. Kunth in Maranhão, BrazilFirst report of mites (Gamasida: Laelapidae) parasitic on wild rodents in Uruguay, with new host records author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Neotropical Entomology

Print version ISSN 1519-566XOn-line version ISSN 1678-8052

Neotrop. Entomol. vol.35 no.5 Londrina Sept./Oct. 2006 



Phenology of Bombus pennsylvanicus sonorus Say (Hymenoptera: Apidae) in Central Mexico


Fenología de Bombus pennsylvanicus sonorus Say (Hymenoptera: Apidae) en el Centro de México



Juan Di Trani de la Hoz

Apartado postal 0830 -7899-29, San Francisco, Panamá,




We studied the seasonal activity of Bombus pennsylvanicus sonorus Say in subtropical conditions of the Mexican Central Plateau. Monthly during 1998, we recorded caste of active individuals (inferred form corporal size measured as wing length), presence and activity of reproductive individuals, and mating activity. Also, we recorded the flower plants used as resources.
Subtropical conditions of the Mexican Central Plateau do not seem to modify phenological features of B. pennsylvanicus sonorus since the species presents annual colonies and a clearly defined period of inactivity. The size of individuals progressively increased between the first recorded activity period in middle April and the end of the colonies in November. Reproductive bees were observed since the second half of the year. The presence of males was recorded between July and November and queens and mating pairs were observed during November and December. Then mating queens were noted seeking hibernation places. Activity resumed in February of the following year. Seasonal activity seems to be more related to availability of floral resources (which, in turn, is related to rain regime), than to changes in temperature and day length. Some of the main food resources used by B. pennsylvanicus sonorus were Tithonia tubiformis, Cosmos bipinnatus, Anoda cristata, Solanum rostratum and Jacaranda mimosaefolia.

Key words: Bumblebee, Fervidobombus, subtropical


Estudiamos la actividad estacional de Bombus pennsylvanicus sonorus Say bajo las condiciones subtropicales de la Meseta Central Mexicana. Mensualmente, durante el año 1998, registramos las castas de abejas activas (infiriéndolo del tamaño corporal medido por la longitud alar), la presencia y actividad de individuos reproductivos y la actividad reproductiva. También se investigaran las plantas florales usadas como recursos alimenticios. Las condiciones Subtropicales de la Meseta Central Mexicana no parecen alterar las características fenológicas de B. pennsylvanicus sonorus, ya que la especie presenta colonias anuales y un periodo bien definido de inactividad. El tamaño de los individuos se incrementó progresivamente desde las primeras actividades observadas, a mediados de abril, hasta el fin de la colonia en noviembre. Las abejas reproductivas fueron registradas a partir de la segunda mitad del año. Los machos fueron observados entre julio y noviembre, y las reinas y parejas apareándose fueron registradas en noviembre y diciembre. Seguidamente las reinas fueron observadas buscando lugares para hibernar. La actitidad fue reasumida a partir de febrero del siguiente año. La actividad estacional parece estar mas relacionada a la disposición de recursos florales (los cuales, a su vez, están relacionados al régimen de lluvias), que a cambios en la temperatura y el largo del día. Algunos de los principales recursos alimenticios de las abejas B. pennsylvanicus sonorus fueron Tithonia tubiformis, Cosmos bipinnatus, Anoda cristata, Solanum rostratum y Jacaranda mimosaefolia.

Palabras clave: Abeja, Fervidobombus, subtropical



Bumblebees, with the exception of the social parasites, produce eusocial annual colonies, founded by a mated female or queen. Under arctic and temperate conditions, colonies are founded during the spring (Plowright & Laverty 1984). The first workers are small individuals, morphologically similar to the queen. Later in the season, the colony produces queens and males, which mate usually with individuals from other colonies. The colony ends after producing reproductive individuals (males and females) and the daughter queens mate at the end of the colony cycle. The newly mated queens hibernate until the following spring. On the other hand, bumblebees of the subgenus Fervidobombus are unusual because they exhibit social characteristics (not usually found in bumblebees) more typical of the highly eusocial honeybees and stingless bees. These include large colonies of up to several thousand individuals with unusually well developed defense systems (Dias 1958, Michener y LaBerge 1954, Cameron et al. 1999), and perennial colonies that can persist for many years.

Despite Bombus pennsylvanicus Degeer being one of the most abundant and widely distributed bumblebee species in North America, there are a few studies on their phenology and behavior (Johnson 1986, Goldblatt & Fell 1987, Todd et al. 1991, Visscher & Vetter 1995).

Two subspecies, B. pennsylvanicus pennsylvanicus and B. pennsylvanicus sonorus, are recognized. The former is more frequently found in South West U.S. and the drier areas of the Mexican Plateau, the Sonoran Desert and Baja California; the second subspecies occurs in the moister areas of the Mexican Plateau, up to the highlands of Oaxaca (Labougle 1990).

The seasonal distribution of the species is partially known for Mexico, based on collection dates from museum specimens (Labougle 1990), and some phenological features of this species were studied by MacFarlane et al. (1994) in Canada.

With this study we investigate the influence of subtropical conditions on the phenology of B. pennsylvanicus sonorus, analyzing the activity and phenology of colonies in Cholula, Puebla, located on the Central Mexican Plateau.


Materials and Methods

Study site. Fieldwork was conducted in three observation points located within and around the campus of the Universidad de las Américas-Puebla, in Cholula, Puebla. The campus occupies 78 ha, and is located at 19°03'13"N 98°16'52"W, at a mean altitude of 2250 m. The climate is classified as C (w2)x', in the system of Köppen with rains predominantly in the summer, and dry and cold winters. The original vegetation was transitional between pine-oak and tropical deciduous forest to the south and southwest, and xerophytic shrub and grasslands to the north and northeast. However, due to intense urban land use and agricultural activities practiced since pre-Columbian times, practically nothing of the original vegetation remains (Rzedowski & Rzedowski 1985).

Sampling. Sampling was performed once a week, between March 26, 1998, when the first active individuals were found, until December 11, 1998, when flying activity was no longer apparent. We captured all bumblebees found in each site during one hour using an entomological net and sacrificed them with Potassium Cyanide. We measured the length of the anterior right wing, from the basal end of the coastal vein, to the apical margin of the wing. Similar measures of wing length have been used for social wasps and were strongly correlated with dry body mass (O'Donnell 1995, O'Donnell 2001).

To estimate the age of individuals (Cartar 1992), the degree of wing wear was semi-quantitatively determined, assigning each individual to a category between 0 (no wear) and 3 (maximum wear observed).

Caste assignation. Female individuals were assigned to either the queen or the worker caste, depending on the measure of wing length. Any individuals with wing length equal to or greater than 1.72 cm were considered queens. Individuals with wing length smaller than 1.72 cm were considered workers. We used this wing length value to differentiate castes because it was the smallest measure obtained during the period when only queens were active. In cases when there was doubt about the caste of an individual, we dissected the reproductive system to determine if the ovaries were developed.

Colony cycle. The colony cycle kind (annual or continuous) was inferred from the presence of individuals in the field and their relative abundance during the observations and samples taken over one year.

First workers emergence. The first workers emergence was determined from the measurements of the anterior wing length of the specimens collected during the observations. The foraging activity of the worker caste began when two groups of individuals were observed, with noticeable different wing length, the queens (bigger bees) and workers (smaller bees). The wing length was chosen because it provides a constant proportion of the size of the insect.

Emergence of reproductive individuals. The production of reproductive individuals of the colony was determined by examining the bees collected. The males usually have a more cylindrical body and fly more slowly. Also, they have an extra antennal segment than females and obvious differences in the reproductive system. Queen-worker differentiation was made as described on "caste assignation".

Mating period. The mating period of the bees was determined from direct observations of individuals in copula in the field and from inferences of the proportion of queen/males during the collections over the course of the year.

Hibernation. The beginning of the hibernation stage was deducted from the decrease in the number of foraging bees and the increase in the size of the active bees. The end of this period lead to the initiation of new queens searching for places to found new colonies.

Plants visited. During our study we collected the most common plant species flowering on the observation sites in which B. pennsylvanicus sonorus visitations were observed. Then, samples were identified by the personal of BUAP herbarium.



The queen caste prevails on the first months of the year. Thus, in March, they represent 100% of the samples, and in the following months (April and May) the proportion decreases gradually until they are absent in the samples at the end of May (Table 1, Fig. 1). In November and December, the queens are again collected in the samples. Moreover, the queens show the lowest variations in wing length throughout the year among the three castes (Fig. 2).







The workers did not appear in the samples until mid April. This represents the first worker generation, characterized by their small bodies (Fig. 3). In the following months, there is a trend of increasing body size of the caste. The wing length average in the first months (April and May) was between 1.3 cm and 1.4 cm, whereas in the lasts months (November and December) it was close to 1.6 cm.



On the other hand, the proportion of workers in the samples reach their maximum in mid year (May to July), when they represent 100% of the castes in the samples (Fig. 1).

The first males emerge in July. Their proportion, however, was low (25%) in the first months of their appearance (September and October), until the last months of the year (November and December) when they represent the majority in the samples (75%). These castes also showed a marked trend of increasing body size over the year. The first males produced (May) averaged 1.4 cm, whereas the males of the last months averaged more than 1.6 cm (Fig. 4).



It was also remarkable the overlap between queens and workers in April and May, when both castes were active (Fig. 1). Another superposition occurred between workers and males from July to December, and workers, males and queens in November and December. These last months are the only period of the year in which all the castes are active.



Our results are congruent with the literature regarding the biology of B. pennsylvanicus sonorus in temperate zones (Johnson 1986, Goldblatt & Fell 1987, Todd et al. 1991). Our study suggests that living in a semitropical zone, as Puebla (Central Mexico), does not notoriously alter the typical phenology of the species. Probably the bees form annual colonies beginning activity in mid or late February and ending in late November or the beginning of December.

As the colony evolved, the size of the bees increased significantly. This phenomenon has been extensively reported (Heinrich 1979, Duchateau 1991, Ribeiro 1994).

The bees reach a maximum wing length in March and April as a result of being queens dominant caste, and they are typically the largest bees of the colony. The samples taken in April show the higgest standard deviation during the year, and it could be attributed to the overlap between queens and the first worker generation. In the following months the wing length average decreases since the number of queens decrease and the workers become the dominant caste.

The most stable period (lower Standard Deviation) was between August and September (Table 1, Fig. 5), when workers, males, and even queens were observed. The characteristic stability from these lapse could be attributed to the size of the colony (number of individuals), which was big enough to produce large quantities of food, in order to support larger bees.



The workers were absent in the samples taken in March because the colonies had just formed and the only foraging bees were queens. Then in April the first workers were collected but their proportion was still low, because the bulk of the foraging activity was still done by the queens. These first workers are characteristically very small bees that represent the first or second generation of the colony. At this stage of the colony the queen is the only foraging bee, so the food supply is typically scarce. In the subsequent months, until July, the workers became the only active foraging caste and their body size increased notoriously. Thus, the last workers produced in the colony were quite large, reaching sometimes the size of the queens. The proportion of workers decreased again when the reproductive individuals appeared, and by the end of the year they had disappeared completely.

The turning point of the colony when the reproductive individuals were produced was between July and August, when the first males were produced. These medium sized males were similar to workers (on wing length), and could not mate because queens were produced a couple of months later. In the subsequent months the proportion of the samples and the male size increased until they represented the majority of the active bees and reached the queens size on November and December.

Queens were the most stable caste in the colony (lesser Standard Deviation on wing length average). It could be a result of queens being produced in a relatively short time frame, in which the food resources of the colony have reached their maximum. The queens were the only surviving individuals of the previous season because they spent the last months of the year hibernating. Then, in February and March they were the only active caste, when they were looking for a location to found a new colony, or foraging for the first worker larva. When these bees completed their development they assisted the queen foraging until they gradually assumed all the foraging work. This explains the decrease of the queens in the samples of the subsequent months: the queens were only dedicated to tasks inside the nest whereas the workers did all the foraging work. It is not until November and December when the queens appear again in the samples. These are new queens, which once emerged, made some foraging trips, mated and found a place for hibernating.

The mating period for B. pennsylvanicus in Cholula seems to be restricted to November and December. This is inferred from just one observation of individuals in copula, and this is the only interval in which there were active males and queens.

Once the queens are fertilized they are buried for hibernation. The length of this period is determined by environmental conditions such as temperature and humidity. Laverty & Plowright (1985) indicate that Bombus of neartic have long hibernation periods, extending for eight months. On the other hand, the tropical species does not hibernate (Michener 1974). Our study showed B. pennsylvanicus in Cholula, a subtropical zone, behave as typical species do in temperate zones. It can be concluded that they have hibernation periods that extend from mid December to mid February, a period of time in which there were not active bees. Finally, the bulk of the food resources used by the species came form annual herbs, native and very common in the mid and lowlands of Mexico Valley (Rzedowski 1985). Some of them were Solanum rostratum, from May to July, and Tithonia tubiformis, Cosmos bipinnatus, Anoda cristata, from August to November. The shape of these flowers indicate they are visited by general pollinators. Also important in the diet of the bees is Jacaranda mimosaefolia, a caducifolium introduced tree from South America, blooming from March to May. From December to February we could not find any flowering plants in the sampling zone. The availability of resources used and environmental factors such as temperature and rain (Fig. 6) seems to determine the cyclic progression of the colony.




To Carlos Vergara, advisor on the thesis from which this article has been originated. To Universidad de las Americas Administration, for allowing the use of the land and facilities throughout the investigation. Also to the personal of the Herbarium of BUAP for the plant identification.



Cameron, S.A., J.B. Whitfield, M. Cohen & N. Thorp. 1999. Novel use of walking trails by the Amazonian bumble bee, Bombus transversalis (Hymenoptera: Apidae). In G.W. Byers, R.H. Hagen & R.W. Brooks (eds.), Entomological contributions in memory of Byron A. Alexander, Univ. Kans. Nat. Hist. Mus. Spec. Publ. 24: 187-193.        [ Links ]

Cartar, R.V. 1992. Morphological senescence and longevity: an experiment relating wing wear and life span in foraging wild bumble bees. J. Anim. Ecol. 61: 225-231.        [ Links ]

Dias, D. 1958. Contribuição para o conhecimento da bionomia de Bombus incarum Franklin da Amazônia (Hymenoptera: Bombidae). Rev. Bras. Entomol. 8: 1-20.        [ Links ]

Duchateau, M.J. 1991. Regulation of colony development in bumblebees. Acta Horticulturae 288: 139-143.        [ Links ]

Goldblatt, J.W. & R.F. Fell. 1987. Adult longevity of workers of the bumble bees Bombus fervidus (F.) and Bombus pennsylvanicus (De Geer) (Hymenoptera: Apidae). Can. J. Zool. 65: 2349-2353.        [ Links ]

Heinrich, B. 1979. Bumblebee economics. Harvard University, Cambridge, Massachusetts, 245p.        [ Links ]

Johnson, R.A. 1986. Intraespecific resource partitioning in the bumble bee Bombus ternarius and B. pennsylvanicus. Ecology 67: 133-138.        [ Links ]

Labougle, J.M. 1990. Bombus of México and Central America (Hymenoptera, Apidae). Univ. Kansas Sci. Bull. 54: 33-73.        [ Links ]

Laverty, T. M. & R.C. Plowright. 1985. Comparative bionomics of temperate and tropical bumble bees with special reference to Bombus ephippiatus (Hymenoptera: Apidae). Can. Entomol. 117: 465-474.        [ Links ]

Macfarlane, R.P., K.D. Patten, L.A. Royce, B.K. Wyatt & D.F. Mayer. 1994. Management potential of sixteen North American bumble bee species. Melanderia 50: 13-19.        [ Links ]

Michener, C. 1974. The social behavior of bees (a comparative study). Harvard University, Cambridge, Massachusetts, 440p.        [ Links ]

Michener, C.D. & W.E. LaBerge. 1954. A large Bombus nest from Mexico. Psyche 61: 63-67.        [ Links ]

O'Donnell, S. 2001. Worker biting interactions and task performance in a swarm-founding eusocial wasp (Polybia occidentalis, Hymenoptera: Vespidae). Behav. Ecol. 12: 353-359.        [ Links ]

O'Donnell, S. & R.L. Jeanne. 1995. The roles of body size and dominance in division of labor among workers of the eusocial wasp Polybia occidentalis. J. Kansas Entomol. Soc. 68: 43-50.        [ Links ]

Plowright, R.C. & T. M. Laverty. 1984. The ecology and sociobiology of bumble bees. Annu. Rev. Entomol. 29: 175-199.        [ Links ]

Rzedowski, J. & G.C. Rzedowski. 1985. Flora fanerogámica del Valle de México. Escuela Nacional de Ciencias Biológicas, México, 674p.        [ Links ]

Ribeiro, M.F. 1994. Growth in bumble bee larvae: Relation between development time, mass and amount of pollen ingested. Can. J. Zool. 72: 1978-1985.        [ Links ]

Todd, E.S., S.L. Buchmann, E.M. Villalobos & M.K. O´Rourke. 1991. Colony ergonomics for a desert-dwelling bumblebee species (Hymenoptera: Apidae). Ecol. Entomol. 16: 361-370.        [ Links ]

Visscher, P.K & R.S. Vetter. 1995. Smoke and target color effects on defensive behavior in yellowjacket wasps and bumble bees (Hymenoptera: Vespidae, Apidae) with description of an electronic attack monitor. J. Econ. Entomol. 88: 579-583.        [ Links ]



Received 26/VIII/05.
Accepted 02/I/06.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License