Creating a �good� science research experience in a liberal

							            “Good” science research experiences in liberal arts colleges.
                                          Caroline Storer


       In 2007, The Chronicle of Higher Education published the article What Good Is

Undergraduate Research, Anyway? Stating that “many students benefit, but studies show

weaknesses in current practices.” In academia, it is almost universally accepted that

undergraduate research provides numerous benefits, increasing student interest in science

disciplines and allowing students to develop skill sets that are advantageous for future job

placement and a career in the field. However, the article in The Chronicle of Higher Education

highlights the recent findings that the advantages of undergraduate research are dependent on the

strength of the mentorship provided not necessarily from the research itself. In the United States,

where over $50 million dollars each year are spent to support nearly 8,000 student undergraduate

research programs are we getting our monies worth?

       At small liberal arts colleges where there are already high demands on a faculty

member’s time and often limited financial resources the increased responsibility of providing a

“good” research experience can be too demanding. I suggest a new model for providing “good”

science education in an undergraduate institution. First, students become involved in research

with both faculty and more senior students during their first year of study. Project leaders would

then be chosen from those students who excelled in this new research atmosphere, taking on a

more demanding role in their second and third years. As a project leader the student would be

expected to teach protocols and the theories behind the protocols new students. They would also

assist graduating students in trouble shooting and project development. In their last years of

study these students would begin acting as project advisors, aiding faculty in developing and

running projects as well as mentoring first year scientists. By having experienced students begin

                                                                                                     1
to take on the role of mentor, lead troubleshooting efforts, and take part in project development a

“good” science education can be provided at a liberal arts institution that produces high quality

student scientists without putting extensive strain on faculty resources.



“Good” Undergraduate Research: A Case Study



       In 2005, Caroline Storer began her first year at Eckerd College and was one of few

selected to participate in the Eckerd College Marine Science Freshman Research Program (FRP).

Through this program she and three other students studied the reproductive ecology of the Gulf

pipefish, Syngnathus scovelli, and were able to present their work at local scientific meeting. The

advantages of beginning research as a freshman were apparent. By being able to include both

independent research and the attendance of a professional meeting on her resume she was able to

secure a summer research internship while only a rising sophomore. Her experience in the FRP

proved to be a catalyst for attaining several other research opportunities and gaining extensive

experience in the field she was studying.

       As a junior, Caroline was able to utilize her research experience as a collabertaor on a

student lead project with her faculty advisor. Together they came up with the idea to study the

molecular population structure of the Gulf pipefish, a study specimen that was easily accessible

in their area, with incoming FRP students. Caroline acted as a project leader for three FRP

students, Rachel Harbeitner, Aisha Rickli-Rahman, and Nathan Van Bibber, setting up the lab,

managing supplies, researching protocols, and most important teaching these students the skills

and the science needed to work in the lab and field.




                                                                                                    2
A synopsis of their work is provided below:

         Molecular population structure and biogeography of the Gulf pipefish in
                                    Florida waters.

       Introduction

               A Gulf of Mexico-Atlantic Ocean division in molecular population

       structure is shared by an array of marine and estuarine species including

       invertebrates such as the American oyster, the horseshoe crab, and the long-

       wristed hermit crab (Young et al. 2002; Avise 1992), bony fishes such as the

       Atlantic sturgeon and a variety of teleosts (Tringali & Bert 1996; Avise 1992),

       and the blacktip shark (Keeney et al. 2005). These species differ greatly in range,

       reproductive strategy, and life history. The similar population structure of these

       independently evolving lineages has been attributed to changes in sea level and

       climate caused by glacial advances and retreats during the Pleistocene epoch

       which may have geographically isolated populations of marine coastal species

       restricted by salinity, temperature, and/or habitat (Avise 1992). Current

       boundaries to dispersal continue to maintain population structure for most of these

       taxa.

               Unlike many of the previously studied species, the Gulf pipefish,

       Syngnathus scovelli, is not restricted along the southeast US coast by salinity or

       high temperature, occupying warm coastal marine to freshwater habitats on the

       Atlantic coast from Georgia to the southern tip of Florida, throughout the Gulf of

       Mexico, and south along Central America to Brazil. However, S. scovelli may

       have limited dispersal capabilities due to its lack of planktonic egg and larval

       stages and its reliance on submerged aquatic vegetation in all life history stages,

                                                                                             3
as evidenced by its absence throughout the West Indies. This combination of

characteristics makes S. scovelli especially interesting from a biogeographical

perspective.



Methods and Materials

       Between 2007 and 2009, 220 pipefish were collected from eight different

locations, three from the Atlantic including St. Johns River (2007), Merritt Island

(2007, 2008), and Fort Pierce (2008), four from the Gulf of Mexico including

Pensacola (2008), Apalachicola/St. Joseph Bays (2007, 2008), Tampa Bay (2007,

2008), and Charlotte Harbor (2007, 2008), and one from the Upper Florida Keys

(2008, 2009). Using DNA isolated from each specimen, a 487 base pair portion of

the mitochondrial DNA control region was amplified in PCR using a protocol and

primers from Teske et al. (2003). Successfully amplified PCR products were

purified and sequenced at the University of Florida Core Sequencing Lab. The

resulting sequences were edited and aligned using ClustalX. Pairwise analysis of

molecular variance (AMOVA) between and within sampling locations and

between years at a single sampling location was calculated using ARLEQUIN

version 2.000 to investigate population structure. To examine the evolutionary

relationship among haplotypes, consensus neighbor-joining and maximum

parsiomony trees from 500 bootstrap replicates were produced in MEGA version

4.0 (Kumar 2008).



Results



                                                                                      4
       Within the 487 bp segment of the mtDNA control region 29 polymorphic

loci were identified for 42 unique haplotypes. Haplotypes 1 through 28 were

found only on the Gulf Coast and in the Florida Keys. These haplotypes are

subsequently referred to as Gulf haplotypes. Haplotype 1 was shared by 49% of

Gulf pipefish and was found at all locations. Haplotypes 29 through 42 were

found in pipefish from only the Atlantic coast. Within the Atlantic, haplotype 29

was most abundant, shared by 61% of individuals found at all three locations. No

haplotypes were shared between Gulf and Atlantic fish.

       There was no genetic difference between years for pipefish sampled

during multiple years at a location (p > 0.202), therefore these fish were grouped

together as one for further analysis. In Florida, S. scovelli was found to be highly

structured as pipefish from all locations on the Gulf coast were significantly

differentiated from fish at all locations on the Atlantic coast. Within the Gulf,

there was no significant genetic difference between Tampa Bay and Charlotte

Harbor (ΦST = -0.01194, p = 0.9272), between Apalachicola/St. Joseph Bays and

Pensacola (ΦST = -0.01288, p = 0.5727) and between Pensacola and the Upper

Florida Keys (ΦST = 0.10475, p = 0.0818). In the Atlantic, there was no

significant genetic difference between Fort Pierce and St. Johns River (ΦST =

0.04017, p = 0.1455). Pairwise genetic distance was not correlated to the

geographical costal distance (R2 = 0.0171, p = 0.5070) between locations.

       Both maximum parsiomony and neighbor-joining trees produced similar

topological structures. The consensus trees show a single genetic lineage of




                                                                                       5
Atlantic haplotypes with a Gulf ancestor. No distinct evolutionary relationships

associated with geographical location appear in either tree.



Discussion

       Genetic differences among some Gulf populations, and particularly,

between the Gulf and Atlantic regions suggest historical limits to dispersal and

gene flow in this species. The existence of a major barrier to gene flow is further

indicated by the lack of a single shared haplotype between the Gulf and Atlantic

populations. The similarity of pipefish in the Upper Keys to those in the Gulf

suggests that this break occurs in SE Florida. This Gulf-Atlantic population

separation in SE Florida is characteristic of other previously-studied marine

species, however the biogeographical history giving rise to this pattern may be

different for S. scovelli. Both maximum parsimony and neighbor joining

haplotype phylogenys suggests an Atlantic clade derived from Gulf relatives. This

could have occurred after the most recent Pleistocene cooling forced the species

into a southern refugium in the southern Gulf of Mexico or the Caribbean. Then,

as the climate warmed, the species may have expanded northward along the Gulf

coast and around Florida, followed by the isolation of the fish that gave rise to

those presently found on the Atlantic coast. This hypothetical re-colonization

scenario is depicted in Figure 1. The absence of S. scovelli in the West Indies

suggests that it is incapable of open-water dispersal and supports this scenario of

re-establishment along a coastal route through the Gulf of Mexico.




                                                                                      6
       Figure 1. Panel depicting the present day distribution of S. scovelli on the far left
       with the hypothetical historical distribution in the middle, and the proposed
       coastal re-colonization route on the far right.


Was this collaborative effort a success?

         All four students involved gained not only experience in the lab, but also experience

leading others in the lab as well. Each student involved took part in preparing and presenting

their project for the annual meeting of the Florida Chapter of the American Fisheries society.

Both the opportunity to prepare for and attend a professional meeting was one of the most

beneficial outcomes of the research as it exposed these students to professional research

community. One FRP student, Rachel, went on to take a summer internship at a state molecular

lab just after her freshman year. In the coming year, Nate, Rachel, and Caroline joined to

together to lead the new group of FRP students in two new projects. Caroline was able to adapt

this initial pipefish research for her senior thesis. All students involved retained their interest in

science research and took on greater responsibilities and roles in projects.



Works Cited

Avise, J.C. 1992. Molecular population structure and the biogeographic history of a regional
       fauna: a case history with lessons for conservation biology. Oikos 63:62-76.
Kenny, D.B., M. Heupel, R.E. Hueter, & E.J. Heist. 2003. Microsatellite and mitochondrial DNA
       analysis of the genetic structure of the blacktip shark (Carcharhinus limbatus) nurseries
       in the Northwestern Atlantic, Gulf of Mexico, and Caribbean Sea. Molecular Ecology
       14:1911-1923.
Kumar, S., M. Nei, J. Dudley, and K. Tamura. 2008. MEGA: A biologist-centric software for
       evolutionary analysis of DNA and protein sequences. Briefing in Bioinformatics 9: 299 -
       306.
Rice, W.R. 1989. Analyzing tables of statistical tests. Evolution 43: 223-225.
Teske, P.R., M.I. Cherry & C.A. Matthee. 2003. Population genetics in the endangered Knysna
       seahorse, Hippocampus capensis. Molecular Ecology 12: 1730-1750.
Tringali, M.D. and T.M. Bert. 1996. The genetic stock structure of common snook Centropomus
       undecimalis. Canadian Journal of Fisheries and Aquatic Sciences 53:974-984.



                                                                                                    7
Young, A.M., C. Torres, J.E. Mack, & C.W. Cunningham. 2002. Morphological and genetic
      evidence for vicariance and refugium in Atlantic and Gulf of Mexico populations of the
      hermit crab Pagurus longicarpus. Marine Biology 140:1059-1066.




                                                                                          8