The mean of 50 observations was 36. It was found later that an observation 48 was wrongly taken as 23. The corrected new mean is :

• 35
• 36.5
• 40
• 42