diff --git a/scrapy/contrib/spiders/crawl.py b/scrapy/contrib/spiders/crawl.py
index d7e42f6f8..7dc3dacd6 100644
--- a/scrapy/contrib/spiders/crawl.py
+++ b/scrapy/contrib/spiders/crawl.py
@@ -86,6 +86,13 @@ class CrawlSpider(Spider):
             rule.process_links = get_method(rule.process_links)
             rule.process_request = get_method(rule.process_request)
 
+    @classmethod
+    def from_crawler(cls, crawler, *args, **kwargs):
+        spider = super(CrawlSpider, cls).from_crawler(crawler, *args, **kwargs)
+        spider._follow_links = crawler.settings.getbool(
+            'CRAWLSPIDER_FOLLOW_LINKS', True)
+        return spider
+
     def set_crawler(self, crawler):
         super(CrawlSpider, self).set_crawler(crawler)
         self._follow_links = crawler.settings.getbool('CRAWLSPIDER_FOLLOW_LINKS', True)
diff --git a/tests/test_spider.py b/tests/test_spider.py
index 53daf39fb..188bef6e4 100644
--- a/tests/test_spider.py
+++ b/tests/test_spider.py
@@ -220,6 +220,30 @@ class CrawlSpiderTest(SpiderTest):
                            'http://example.org/about.html',
                            'http://example.org/nofollow.html'])
 
+    def test_follow_links_attribute_population(self):
+        crawler = get_crawler()
+        spider = self.spider_class.from_crawler(crawler, 'example.com')
+        self.assertTrue(hasattr(spider, '_follow_links'))
+        self.assertTrue(spider._follow_links)
+
+        crawler.settings.set('CRAWLSPIDER_FOLLOW_LINKS', False)
+        spider = self.spider_class.from_crawler(crawler, 'example.com')
+        self.assertTrue(hasattr(spider, '_follow_links'))
+        self.assertFalse(spider._follow_links)
+
+    def test_follow_links_attribute_deprecated_population(self):
+        spider = self.spider_class('example.com')
+        self.assertFalse(hasattr(spider, '_follow_links'))
+
+        spider.set_crawler(get_crawler())
+        self.assertTrue(hasattr(spider, '_follow_links'))
+        self.assertTrue(spider._follow_links)
+
+        spider = self.spider_class('example.com')
+        spider.set_crawler(get_crawler({'CRAWLSPIDER_FOLLOW_LINKS': False}))
+        self.assertTrue(hasattr(spider, '_follow_links'))
+        self.assertFalse(spider._follow_links)
+
 
 class SitemapSpiderTest(SpiderTest):