Tuta absoluta Meyrick (Lepidoptera: Gelechiidae), the South American tomato leafminer, is an insect pest that causes major losses to tomato (Solanum lycopersicum L.; Solanaceae) production worldwide. Up until the 2000s, its effects were limited to its native range in South America. Within that range, T. absoluta proved difficult to control by chemical means because not only do their larvae gain shelter from pesticides by mining inside tomato leaves, stems, and fruits (Tropea Garzia et al. 2012), but T. absoluta populations developed resistance to a range of pesticide classes (Ferracini et al. 2012; Guedes & Picanço 2012). Pesticide resistance, coupled with a high rate of reproduction (up to 11 generations per yr), facilitate T. absoluta outbreaks that can destroy up to 100% of infested tomato crops (EPPO 2005; Desneux et al. 2011). Tuta absoluta attained global prominence as an invasive pest in 2006 when it was introduced into Spain (Tropea Garzia et al. 2012). It has since invaded other countries in Europe, the Mediterranean, northern Africa (Desneux et al. 2011; Mohamed et al. 2012), and South Asia (Kalleshwaraswamy et al. 2015; Shashank et al. 2015; Bajracharya et al. 2016; Hossain et al. 2016). Tuta absoluta also crossed the Sahara Desert and is now seen as a serious threat to tomato production in sub-saharan Africa (Brévault et al. 2014), where preliminary reports suggest it has already reached South Africa, invading Tanzania and several other countries on the way (Chidege et al. 2016; Campos et al. 2017).
The presence of Tuta absoluta in Tanzania, which was first documented by the Tropical Pesticides Research Institute in 2014 (Chidege et al. 2016), stands as a major threat to tomato production, which annually exceeds the weight of Tanzania's other horticultural products combined (Minja et al. 2011). Although the extent of T. absoluta damage in Tanzania has not been rigorously estimated, it is understood that poorly managed T. absoluta infestations lead to devastating reductions of tomato yields (Desneux et al. 2010; Guedes & Picanço 2012). Indeed, the severity of the infestation in Tanzania attracted significant media attention (Ihucha 2015, 2016). Research is needed to assess the scope of the problem across Tanzania's major tomato growing regions and thereby inform local and international agencies seeking to respond with funding, research, and grower education.
Research also is needed to determine the host range of T. absoluta in Tanzania. Although T. absoluta prefers tomato as its host (EPPO 2005), it also attacks a number of other of solanaceous crops including eggplant (Solanum melongena L., also called aubergine), potato (Solanum tuberosum L.), tobacco (Nicotiana tabacum L.), and African eggplant (Solanum aethiopicum L.) (EPPO 2005; Brévault et al. 2014; Mohamed et al. 2015). Tuta absoluta also has been found on a range of wild solanaceous plants (e.g., Solanum spp. and Datura spp.), which indicates that weed sanitation may be an important tactic for managing pest populations (EPPO 2005; Mohamed et al. 2015). Furthermore, 1 of the commonly cited wild hosts, Solanum nigrum L., is a cultivated vegetable (African nightshade) in Tanzania and may be at risk. It is important to note that different T. absoluta populations may vary in host usage. For example, a study in Argentina found that T. absoluta caterpillars could not survive on eggplant, whereas a study in Sudan reported that they could survive on it (Pereyra & Sánchez 2006; Mohamed et al. 2015). It is possible that invasive populations of T. absoluta are adapting to novel hosts (Abbes et al. 2012; Mohamed et al. 2015), because besides Solanaceae, T. absoluta has been reported to feed on plants from the families Amaranthaceae, Asteraceae, Brassicaceae, Convolvulaceae, Cucurbitaceae, Euphorbiaceae, Fabaceae, and Poaceae (Mohamed et al. 2015; CABI 2017). Thus, it is important to determine in which hosts T. absoluta is able to develop in each region that it invades.
In this study, we report on a survey conducted in 2015 to determine the host range and infestation rates of T. absoluta on solanaceous crops and common weeds in Tanzania. The survey was conducted in major tomato producing regions of Tanzania.
Materials and Methods
SELECTION OF SUR VEY SITES
Four leading tomato growing districts of Tanzania were identified in consultation with the Tanzanian Horticultural Association: Arumeru District of Arusha Region, Lushoto District of Tanga Region, Kilolo District of Iringa Region, and Mvomero District of Morogoro Region (Fig. 1). The District Agricultural Officers from each of these districts were contacted to obtain lists of 10 to 20 villages where tomato and at least 2 other solanaceous crops were being cultivated. The lists of villages were randomly ranked and the lowest-ranking villages (numbers 1, 2, and 3) were pre-selected as survey sites. The Agricultural Extension Agents from each location were contacted to arrange visits. In cases where the extension agent was not available, the next-lowest ranked village was substituted. Also, where geographic spacing of the 3 randomly selected villages was less than 10 km apart, 1 of the villages was exchanged for the next lowest ranking village. However, in one instance, in the Arumeru District, 2 villages that we surveyed were closer than expected (3.4 km apart).
The aforementioned sites were surveyed between 9 and 30 Sep 2015. Because none of the villages that we initially surveyed in Arumeru were growing potato, 3 additional Arumeru villages were visited to survey in potato in mid-Oct 2015. The survey locations, as well as crops and weeds at each location, are summarized in Table 1. Our sampling activities targeted the dry season before the short Nov/Dec rainy season.
Survey maps were produced with the statistical software R in conjunction with software packages ggmap, maptools, rgdal, and ggplot2 (Wickham 2009; Bivand et al. 2012; Kahle & Wickham 2013; R Core Team 2014; Bivand & Lewin-Koh 2015). District outlines were extracted from administrative maps from DIVA-GIS (Hijmans et al. 2012).
Within each randomly selected village, we sampled 1 field per available solanaceous crop. Fields were pre-selected by the Agricultural Extension Officer and typically represented the nearest fields to our meeting location. Each crop field was sampled in a zigzag pattern at 10 points that included interior plants as well as plants on the field margins; however, the extreme margins (2 m border) were not sampled. At each sample point we counted the total leaves and total T. absoluta mines from 1 individual plant so as to determine the number of mines per leaf. Additionally, at each point the 10 nearest fruits (if available) were sampled for any signs of T. absoluta damage, for a total of 100 fruits per field. To confirm T. absoluta as the agent of observed fruit damage, we looked for the characteristic pin-hole damage caused by T. absoluta caterpillars in tomato, or we dissected the fruit with a knife (other fruits). Tuta absoluta caterpillars were identified in the field according to 3 criteria: characteristic leaf mines, body form and size as described in USDA (2011), and a distinctive dark prothoracic band as described in Roditakis et al. (2010).
Overall, the survey included the following crops: tomato, African eggplant, eggplant (aubergine), pepper (Capsicum annuum L.), African nightshades (Solanum americanum Mill. and Solanum nigrum L.), and potato. We also investigated whether weeds harbor T. absoluta larvae by sampling weedy solanaceous species around T. absolutainfested tomato fields in Arumeru District. Because weed distribution was patchy rather than uniform throughout the fields, we haphazardly selected 10 plants from inside and around tomato fields. Species surveyed included Datura stramonium L. var inermis (Jacq.), Solanum incanum L., and Nicandra physalodes (L.) Gaertn. No other solanaceous weeds were observed in the vicinity of tomato fields in the Arumeru District (the district in which target weeds were identified for this survey).
Representative weed specimens are archived in the National Herbarium of Tanzania. Caterpillars from T. absoluta-like mines in weed leaves were collected alive and reared to adulthood in plastic containers in the laboratory. Caterpillars from S. incanum did not survive on S. incanum leaves nor on tomato leaves and, thus, could not be identified. Caterpillars from D. stramonium and N. physalodes survived well on their respective host leaves. Emerging adults were sent to Dr. Sangmi Li (Arizona State University) for identification and archived in the Arizona State University Hasbrouck Insect Collection.
FARMER SUR VEY
When available, growers from each location were asked whether they recognized T. absoluta (either by name or by photo), and what crops, if any, they thought that T. absoluta affected. They also were asked what interventions they had taken for T. absoluta (if any). Only 3 to 5 growers per district were surveyed because others were unavailable to comment.
GEOGRAPHIC RANGE OF TUTA ABSOLUTA IN TANZAN IA
HOST RANGE AND INFESTATION RATES OF TUTA ABSOLUTA
Among the crops and weeds that we surveyed, tomato, eggplant (aubergine), African nightshades, and potato all exhibited T. absoluta mines, but African eggplant and pepper did not (Fig. 3A; Table 2). Only tomato sustained damage to the fruit from T. absoluta caterpillars (Fig. 3B; Table 3). In tomato fields, mine density ranged from 0.1 to 4.4 mines per leaf, and the percentage of fruits damaged by T. absoluta ranged from 0 to 41%. In eggplant fields, T. absoluta damage varied from 0 to 2.7 mines per leaf, but fruits were not affected by T. absoluta. In potato fields T. absoluta damage ranged from 0 to 0.3 mines per leaf. In African nightshades, T. absoluta damage ranged from 0 to only 0.08 mines per leaf. Average foliage and fruit damage levels for each crop and district are presented in Tables 2 and 3. No T. absoluta mines were detected in tissues or fruits of pepper plants or African eggplants.
Survey locations and crops sampled at each site in Tanzania. Each “X” represents an individual field. Not all crops were available in all villages.
Although we observed leaf mines in D. stramonium and N. physalodes that resembled T. absoluta feeding galleries, caterpillars reared from these mines to adulthood proved not to be T. absoluta, but Phthorimaea operculella Zeller (Lepidoptera: Gelechiidae), the potato tuberworm. We also observed feeding galleries in S. incanum leaves that superficially resembled T. absoluta galleries, although in these samples the caterpillars appeared to have consumed the lower leaf surface (which is usually not consumed by T. absoluta), leaving only the dense trichome layer of the plant to serve as the lower boundary of the mine. Caterpillars from S. incanum mines that we attempted to rear in the laboratory died before pupating, but we concluded that they likely were not T. absoluta because they refused to eat tomato leaves and they lacked the distinctive black band behind the head that is characteristic of T. absoluta.
GROWER AWARENESS OF TUTA ABSOLUTA
A majority of growers (15 out of 19) had heard of T. absoluta prior to our survey, and 16 out of 19 recognized photos of the moth whether or not they knew its name. Although 15 of the growers knew that T. absoluta could affect tomato crops, only 1 suspected that it also affects eggplant and nightshades, and none of the farmers answered that T. absoluta affects potato. The only pest management strategy that these farmers knowingly employed against T. absoluta was spraying conventional pesticides.
This survey confirms the recent report of T. absoluta in Tanzania (Chidege et al. 2016), and establishes that T. absoluta has invaded each of the major tomato producing regions of Tanzania: Arusha, Tanga, Morogoro, and Iringa. The presence of T. absoluta in Tanzania poses a serious threat to tomato production. Although most growers are keeping fruit damage well below 50% with the assistance of conventional pesticides, net yield losses are actually greater than the observed fruit damage because foliage damage also reduces tomato yields by limiting photosynthetic capacity of the plants. This challenge for growers is likely to intensify given the propensity of T. absoluta to evolve resistance to conventional pesticides (Guedes & Picanço 2012). In general, evolution of pesticide resistance may be delayed through spray rotation (Silva et al. 2011) and integration with other tactics as part of an integrated pest management or IPM strategy, but the farmers we surveyed were relying on conventional pesticides as their sole management tactic, and to our knowledge pesticide rotation is not a common practice in Tanzania or other countries in the region. Although the farmer survey sample size is admittedly small at only 19 (because our surveillance methods did not require growers to be present during surveys), it suggests that most tomato growers in Tanzania's leading tomato production regions recognize T. absoluta as a tomato pest, but they do not approach management with an IPM paradigm. Overall, at least 15% of tomatoes are damaged, and the true value is likely to be higher because our survey included tomatoes that would be on the vine and exposed to T. absoluta for additional weeks.
Although T. absoluta targeted tomato as a host more than other solanaceous crops (Fig. 3), tomato is not the only crop susceptible to T. absoluta infestations. Eggplant (aubergine), potato, and African nightshades also frequently sustained foliar damage from this pest in our study (Fig. 3). These findings concur with reports from Senegal, Sudan, and Tunisia regarding the susceptibility of these crops to T. absoluta (Desneux et al. 2010; Brévault et al. 2014; Mohamed et al. 2015; Abbes et al. 2016). But unlike Mohamed et al. (2015), we did not observe any damage to eggplant (aubergine) fruits. This suggests that reductions to eggplant yields in Tanzania are thus far indirect, although the possibility of direct fruit damage by T. absoluta in the future cannot be ruled out. By contrast, damage to edible parts of African nightshade was direct, because nightshades are grown for their leaves. Whereas nightshades (S. nigrum) are regarded merely as weeds in other parts of the global distribution of T. absoluta (Desneux et al. 2010), they are important vegetable crops in East Africa, which heightens the local significance of this pest-host association. Only a small fraction, less than 1%, of nightshade leaves were infested in the fields that we surveyed. However, one of the authors of the current study (S. T.) has observed T. absoluta causing much greater harm to nightshade crops in Manyara Region (data not shown). These differences in infestation rates may indicate useful variation in the susceptibility of cultivated varieties of nightshades that could be exploited to minimize future losses. Alternatively, the differences in infestation rates may be due to the relative distribution or absence of other major host plants of T. absoluta.
Two crops that were not affected by T. absoluta are pepper and African eggplant. These crops are sometimes omitted from lists of T. absoluta host plants (e.g., EPPO 2005; Desneux et al. 2010). However, recent reports indicate that T. absoluta infested pepper in Turkey (Bayram et al. 2015) and African eggplant in Senegal (Brévault et al. 2014). Despite these reports, our survey indicates that these crops are not currently affected by T. absoluta in Tanzania.
Tuta absoluta mines per leaf in 9 solanaceous species in 4 districts of Tanzania (mean ± SD between fields; number of fields shown in parentheses). Ten plants per field were sampled in a zigzag pattern.
Percentage of fruits damaged by Tuta absoluta in 4 districts of Tanzania (mean ± SD between fields; number of fields shown in parentheses). In each field, 100 fruits in total were sampled from 10 locations in a zigzag pattern. No fruits were sampled from potato (because they are cultivated for tubers, not fruits).
In addition to crop plants, we examined whether wild solanaceous plants in and around tomato fields in Arumeru might be serving as alternate hosts for T. absoluta. Although we identified 3 weeds bearing T. absoluta-like mines inhabited by lepidopteran larvae (S. incanum, D. stramonium, and N. physalodes), our follow-up work established that none of the caterpillars were T. absoluta. These data can be regarded as evidence that T. absoluta is not using common solanaceous weeds as hosts; however, it should be considered preliminary because only 3 sites were formally surveyed for D. stramonium, and only 1 site each for S. incanum and N. physalodes (Table 2). That said, we informally looked for T. absoluta mines in these weeds throughout the entire survey and did not find a single T. absoluta-infested weed, which implies that T. absoluta is passing by these plants in favor of tomato and, to a lesser extent, eggplant, potato, and nightshades. However, this does not rule out the possibility that T. absoluta might use non-preferred hosts in the absence of its preferred hosts, because our survey centered on areas that predominantly produce tomato.
Our encounter with P. operculella caterpillars in D. stramonium and N. physalodes when we were looking for T. absoluta is not surprising because these related lepidopterans look similar and use similar hosts (USDA 2011). Yet this experience underscores the importance of confirming larval identity and host compatibility through laboratory studies that complement field surveys. Taking these steps would enhance the usefulness of reports claiming many new host species for T. absoluta (Bayram et al. 2015; Mohamed et al. 2015). Recent examples of such laboratory studies confirmed that eggplant (aubergine) and S. nigrum are viable hosts for T. absoluta, whereas D. stramonium and Datura ferox L. (Solanaceae) are not compatible hosts (Mohamed et al. 2015; Abbes et al. 2016). This later finding was consistent with the absence of T. absoluta on D. stramonium in our study.
Phthorimaea operculella, first detected in Tanzania in the late 1980s by pheromone trapping, is an important pest of potatoes in East Africa (Parker & Hunt 1989). Our findings indicate that P. operculella uses D. stramonium and N. physalodes as alternate host plants. These findings have been observed independently in India and Rhodesia, respectively (Das & Raman 1994). Thus, although our data do not support the practice of removing D. stramonium and N. physalodes for T. absoluta control, sanitation of these weeds still should be recommended in potato IPM programs.
We would like to thank Dr. Sangmi Li (Arizona State University, Hasbrouck Insect Collection) for assistance in identifying the P. operculella specimens. Funding for this research was provided by a Fulbright US Student Award to JDS from the United States Institute for International Education, and by an Innovations Fund Grant to JDS, TD, and SR from the World Vegetable Center.
- Abbes K, Harbi A, Chermiti B. 2012. The tomato leafminer Tuta absoluta (Meyrick) in Tunisia: current status and management strategies.EPPO Bulletin42: 226–233. Google Scholar
- Abbes K, Harbi A, Elimem M, Hafsi A, Chermiti B. 2016. Bioassay of three solanaceous weeds as alternative hosts for the invasive tomato leafminer Tuta absoluta (Lepidoptera: Gelechiidae) and insights on their carryover potential.African Entomology24: 334–342. Google Scholar
- Bajracharya ASR, Mainali RP, Bhat B, Bista S, Shashank PR, and Meshram NM. 2016. The first record of South American tomato leaf miner, Tuta absoluta (Meyrick 1917) (Lepidoptera: Gelechiidae) in Nepal.Journal of Entomology and Zoological Studies4: 1359–1363. Google Scholar
- Bayram Y, Büyük M, ÖzaslanC, BektaşO, Bayram N, Mutlu C, and Bükün B. 2015. New host plants of Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in Turkey.Journal of Tekirdag Agricultural Faculty12: 43–46. Google Scholar
- Bivand R, Keitt T, Rowlingson B. 2012. rgdal: bindings for the geospatial data abstraction library. R package http://CRAN.R-project.org/package=rgdal (last accessed 20 June 2018). Google Scholar
- Bivand R, Lewin-Koh N. 2015. maptools: tools for reading and handling spatial objects. R package http://CRAN.R-project.org/package=maptools (last accessed 20 June 2018). Google Scholar
- Brévault T, Sylla S, Diatte M, Bernadas G, and Diarra K. 2014. Tuta absoluta Meyrick (Lepidoptera: Gelechiidae): a new threat to tomato production in Sub-Saharan Africa.African Entomology22: 441–444. Google Scholar
- CABI (Center for Agriculture and Bioscience International). 2017. Tuta absoluta (tomato leafminer). In Invasive Species Compendium. Center for Agriculture and Bioscience International, Wallingford, United Kingdom. Google Scholar
- Campos MR, Biondi A, Adiga A, Guedes RNC, Desneux N. 2017. From the Western Palaearctic region to beyond: Tuta absoluta 10 years after invading Europe.Journal of Pest Science90: 787–796. Google Scholar
- Chidege M, Al-zaidi S, Hassan N, Julie A, Kaaya E, Mrogoro S. 2016. First record of tomato leaf miner Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in Tanzania.Agriculture and Food Security5: 1–7. Google Scholar
- Das GP, Raman KV. 1994. Alternate hosts of the potato tuber moth, Phthorimaea operculella (Zeller).Crop Protection13: 83–86. Google Scholar
- Desneux N, Luna MG, Guillemaud T, Urbaneja A. 2011. The invasive South American tomato pinworm, Tuta absoluta, continues to spread in Afro-Eurasia and beyond: the new threat to tomato world production.Journal of Pest Science84: 403–408. Google Scholar
- Desneux N, Wajnberg E, Wyckhuys KAG, Burgio G, Arpaia S, Narváez-Vasquez CA, González-Cabrera J, Catalán Ruescas D, Tabone E, Frandon J, Pizzol J, Poncet C, Cabello T, Urbaneja A. 2010. Biological invasion of European tomato crops by Tuta absoluta: ecology, geographic expansion and prospects for biological control.Journal of Pest Science83: 197–215. Google Scholar
- EPPO (European and Mediterranean Plant Protection Organization).2005. Data sheets on quarantine pests: Tuta absoluta.EPPO Bulletin35: 434–435. Google Scholar
- Ferracini C, Ingegno BL, Navone P, Ferrari E, Mosti M, Tavella L, Alma A. 2012. Adaptation of indigenous larval parasitoids to Tuta absoluta (Lepidoptera: Gelechiidae) in Italy.Journal of Economic Entomology105: 1311–1319. Google Scholar
- Guedes RNC, Picanço MC. 2012. The tomato borer Tuta absoluta in South America: pest status, management and insecticide resistance.EPPO Bulletin42: 211–216. Google Scholar
- Hijmans RJ, Guarino L, Mather P. 2012. DIVA-GIS Version 7.5 Manual. DIVA-GIS, http://www.diva-gis.org/docs/DIVA-GIS_manual_7.pdf (last accessed 20 June 2018). Google Scholar
- Hossain MS, Mian MY, Muniappan R. 2016. First record of Tuta absoluta (Lepidoptera: Gelechiidae) from Bangladesh.Journal of Agricultural and Urban Entomology32: 101–105. Google Scholar
- Ihucha A. 2015. Brokers cash in on impending tomato shortage. The Citizen. http://www.thecitizen.co.tz/News/Business/Brokers-cash-in-on-impending-tomatoshortage/1840414-2705608-icvojoz/index.html (last accessed 20 Jun 2018). Google Scholar
- Ihucha A. 2016. Z'bar tomato prices up as pest destroys crop. The Citizen. http://www.thecitizen.co.tz/News/Business/Z-bar-tomato-prices-up-aspest-destroys-crop/1840414-3072224-i0jdckz/index.html (last accessed 20 Jun 2018). Google Scholar
- Kalleshwaraswamy CM, Murthy MS, Viraktamath CA, Kumar NKK. 2015. Occurrence of Tuta absoluta (Lepidoptera: Gelechiidae) in the Malnad and Hyderabad-Karnataka Regions of Karnataka, India.Florida Entomologist98: 970–971. Google Scholar
- Kahle D, Wickham H. 2013. ggmap: spatial visualization with ggplot2.R Journal5: 144–161. Google Scholar
- Minja RR, Ndee A, Swai IS, Ojiewo CO. 2011. Promising improved tomato varieties for eastern Tanzania.African Journal of Horticultural Science4: 24–30. Google Scholar
- Mohamed ESI, Mahmoud MEE, Elhaj MAM, Mohamed SA, Ekesi S. 2015. Host plants record for tomato leaf miner Tuta absoluta (Meyrick) in Sudan.EPPO Bulletin45: 108–111. Google Scholar
- Mohamed ESI, Mohamed ME, Gamiel SA. 2012. First record of the tomato leafminer, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae) in Sudan.EPPO Bulletin42: 325–327. Google Scholar
- Parker BL, Hunt G. 1989. Phthorimaea operculella (Zell.), the potato tuber moth: new locality records for East Africa.American Potato Journal66: 583–586. Google Scholar
- Pereyra PC, Sánchez NE. 2006. Effect of two solanaceous plants on developmental and population parameters of the tomato leaf miner, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae).Neotropical Entomology35: 671–676. Google Scholar
- R Core Team. 2014. R: a language and environment for statistical computing.R Foundation for Statistical Computing, Vienna, Austria. http://www.Rproject.org Google Scholar
- Roditakis E, Papachristos D, Roditakis NE. 2010. Current status of the tomato leafminer Tuta absoluta in Greece.EPPO Bulletin40: 163–166. Google Scholar
- Shashank PR, Chandrashekar K, Meshram NM, Sreedevi K. 2015. Occurrence of Tuta absoluta (Lepidoptera: Gelechiidae) an invasive pest from India.Indian Journal of Entomology77: 323–329. Google Scholar
- Silva GA, Picanco MC, Bacci L, Crespo ALB, Rosado JF, Guedes RNC. 2011. Control failure likelihood and spatial dependence of insecticide resistance in the tomato pinworm, Tuta absoluta.Pest Management Science67: 913–920. Google Scholar
- Tropea Garzia G, Siscaro G, Biondi A, Zappalà L. 2012. Tuta absoluta, a South American pest of tomato now in the EPPO region: biology, distribution and damage.EPPO Bulletin42: 205–210. Google Scholar
- USDA (United States Department of Agriculture). 2011. New pest response guidelines: tomato leafminer (Tuta absoluta). Animal and Plant Health Inspection Service, United States Department of Agriculture, http://www.aphis.usda.gov/import_export/plants/manuals/emergency/downloads/Tuta-absoluta.pdf (last accessed 20 June 2018). Google Scholar
- Wickham H. 2009. ggplot2: Elegant Graphics for Data Analysis, 2nd ed.Springer-Verlag, New York, New York, USA. Google Scholar