diff --git a/modules/societegenerale/sgpe/browser.py b/modules/societegenerale/sgpe/browser.py
index 98a5a1ea..87c42d85 100644
--- a/modules/societegenerale/sgpe/browser.py
+++ b/modules/societegenerale/sgpe/browser.py
@@ -21,7 +21,7 @@
 from weboob.tools.browser import BaseBrowser, BrowserIncorrectPassword
 from weboob.tools.ordereddict import OrderedDict
 
-from .pages import LoginPage, AccountsPage
+from .pages import LoginPage, AccountsPage, HistoryPage
 
 
 __all__ = ['SGProfessionalBrowser', 'SGEnterpriseBrowser']
@@ -34,6 +34,7 @@ class SGPEBrowser(BaseBrowser):
     def __init__(self, *args, **kwargs):
         self.PAGES = OrderedDict((
             ('%s://%s/Pgn/.+PageID=SoldeV3&.+' % (self.PROTOCOL, self.DOMAIN), AccountsPage),
+            ('%s://%s/Pgn/.+PageID=ReleveCompteV3&.+' % (self.PROTOCOL, self.DOMAIN), HistoryPage),
             ('%s://%s/' % (self.PROTOCOL, self.DOMAIN), LoginPage),
         ))
         BaseBrowser.__init__(self, *args, **kwargs)
@@ -67,12 +68,26 @@ class SGPEBrowser(BaseBrowser):
     def accounts(self):
         self.location('/Pgn/NavigationServlet?PageID=SoldeV3&MenuID=%s&Classeur=1&NumeroPage=1' % self.MENUID)
 
+    def history(self, _id):
+        self.location('/Pgn/NavigationServlet?PageID=ReleveCompteV3&MenuID=%s&Classeur=1&Rib=%s&NumeroPage=1' % (self.MENUID, _id))
+
     def get_accounts_list(self):
         if not self.is_on_page(AccountsPage):
             self.accounts()
         assert self.is_on_page(AccountsPage)
         return self.page.get_list()
 
+    def get_account(self, _id):
+        for a in self.get_accounts_list():
+            if a.id == _id:
+                yield a
+
+    def iter_history(self, account):
+        self.history(account.id)
+        assert self.is_on_page(HistoryPage)
+        for transaction in self.page.iter_transactions(account):
+            yield transaction
+
 
 class SGProfessionalBrowser(SGPEBrowser):
     DOMAIN = 'professionnels.secure.societegenerale.fr'
diff --git a/modules/societegenerale/sgpe/pages.py b/modules/societegenerale/sgpe/pages.py
index e2ccdae5..3ec61c72 100644
--- a/modules/societegenerale/sgpe/pages.py
+++ b/modules/societegenerale/sgpe/pages.py
@@ -20,6 +20,7 @@
 from logging import error
 import re
 from decimal import Decimal
+from datetime import datetime
 
 from weboob.tools.browser import BasePage
 from weboob.tools.json import json
@@ -35,6 +36,10 @@ from ..captcha import Captcha, TileError
 __all__ = ['LoginPage', 'AccountsPage']
 
 
+class Transaction(FrenchTransaction):
+    _coming = False
+
+
 class SGPEPage(BasePage):
     def get_error(self):
         err = self.document.getroot().cssselect('div.ngo_mire_reco_message') \
@@ -97,6 +102,30 @@ class AccountsPage(SGPEPage):
                 account.label = to_unicode(tdname)
                 account.id = to_unicode(tdid.replace(u'\xa0', '').replace(' ', ''))
                 account._agency = to_unicode(tdagency)
-                account.balance = Decimal(FrenchTransaction.clean_amount(tdbalance))
+                account.balance = Decimal(Transaction.clean_amount(tdbalance))
                 account.currency = account.get_currency(tdbalance)
                 yield account
+
+
+class HistoryPage(SGPEPage):
+    def iter_transactions(self, account):
+        table = self.parser.select(self.document.getroot(), '#tab-corps', 1)
+        for i, tr in enumerate(self.parser.select(table, 'tr', 'many')):
+            tddate, tdlabel, tddebit, tdcredit, tdval, tdbal = [td.text_content().strip()
+                                                                for td
+                                                                in self.parser.select(tr, 'td', 4)]
+            tdamount = tddebit or tdcredit
+            # not sure it has empty rows like AccountsPage, but check anyway
+            if all((tddate, tdlabel, tdamount)):
+                t = Transaction(i)
+                t.set_amount(tdamount)
+                date = datetime.strptime(tddate, '%d/%m/%Y')
+                val = datetime.strptime(tdval, '%d/%m/%Y')
+                # so that first line is separated by parse()
+                # also clean up tabs, spaces, etc.
+                l1, _, l2 = tdlabel.partition('\n')
+                l1 = ' '.join(l1.split())
+                l2 = ' '.join(l2.split())
+                t.parse(date, l1 + '  ' + l2)
+                t._val = val  # FIXME is it rdate? date?
+                yield t