Recently, electronic book(e-book) market is growing rapidly and people become able to choice e-books that they would like to read from a large amount of e-books. Therefore, systems for finding efficiently one or more sufficient books that have something worth reading from vast numbers of e-books are demanded. In order to support users to select books, many techniques for searching and recommending books have been proposed. However the users would have to judge whether each book in candidates is worth reading. We think that stand reading is effective for the selecting actual books at bookstores in the real world. In this paper, we introduce a method for supporting stand reading on e-books on the Web. Our method recommends a user sentences which would attract and/or interest the user in a book. In our method, firstly, the attractiveness of each term in a book is calculated based on reviews about the book on the Web. Then, the attractiveness of each sentence in the book is calculated based on the attractiveness of the terms. Furthermore, this paper shows the experimental results of our method and discusses its effectiveness.