It is weird, but there is something else going on. It is not an
inherent BufferedReader behavior. Try running this program:

import java.io.BufferedReader;
import java.io.IOException;
import java.io.StringReader;
import java.util.StringTokenizer;

public class BufferedReaderTest {
  public static void main(String[] args) throws IOException {
    // Build a string with a couple of lines, each with lots of
    // tokens
    StringBuffer sb = new StringBuffer();
    for (int i = 0; i < 100; i++) {
      sb.append("a, 9, ");
    }
    sb.append(System.getProperty("line.separator"));
    sb.append(sb.toString());
    String data = sb.toString();

    // Attach a buffered reader to it
    BufferedReader reader = new BufferedReader(new StringReader(
        data));
    System.out.println("Line 1:");
    String line1 = reader.readLine();
    System.out.println(line1);
    System.out.println("Line 2:");
    System.out.println(reader.readLine());

    // Check the number of tokens
    System.out.print("Line 1 tokens = ");
    System.out.println(new StringTokenizer(line1).countTokens());
  }
}

BufferedReader reads a line that StringTokenizer thinks has 200 tokens.

Patricia