In the past, I along with most of my colleagues, would recommend the excellent Fowler, Cohen & Jarvis (Practical Statistics for Field Biologists, OUP) to undergraduate students grappling with basic statistical analysis. However the ubiquity of Excel and the growing popularity of the free but powerful statistical package R make that book look a bit dated today.
Statistics for Ecologists using R and Excel is an excellent replacement covering most of the standard tests I would expect an ecologist or marine biology student at the Centre for Environmental and Marine Sciences to be able to deal with by the end of their 1st year. The reader is led gently through each test in both Excel and R with the assistance of screen shots of relevant examples. I particularly like the emphasis on the basics (planning, data recording, exploring data (looking at numbers and graphs) and reporting results. There is some particularly good advice on transforming data. Most of the tests you expect to see are covered: t-tests, correlation, ANOVAs, regression and the non-parametric equivalents. The author is plainly an ecologist with lots of teaching experience rather than a mathematician! Another feature that I like is the fact that you don't have to flip backwards and forwards through the pages to attempt each test - all of the information you need for each one is in each section. The only criticism I have of the book is that Multi Variate Analyses are not mentioned.
I recommend this text to any undergraduate or tutor that teaches statistical analyses to environmental scientists, field biologists, ecologists or marine biologists.