The aims of the present experiment were to study the effect of grain type and processing on chewing activity in horses. Three adult trotters (Exp.I) and 3 adult Icelandic horses (Exp.II) were fed 3 daily meals during 3 consecutive days in two 3 x 3 completely randomized block design experiments. Meals of (Exp.I: 1.0 kg) and (Exp.II: 0.5 kg) oats, barley and wheat were fed whole, rolled and ground at 10 am, 12 pm, 2 pm. Jaw movements (JM) were identified from pressure oscillations in a tube around the mouth. The efficient chewing time (EPTIME, min/kg DM), corrected for pauses, the basic chewing rate (PBCR, JM/s) and the standard deviation of time interval between JM (SPDDT), which indicate chewing regularity, were estimated. The mean EPTIME was 24 and 15 min for the Icelandic and trotter horses, respectively. The EPTIME for whole grain was 20 min and shorter than for ground grain (P = 0.02). EPTIME for oats was 25 min and longer than for wheat (P=0.05). The SPDDT value for whole grain was low (0.06) and therefore more regularly chewed than ground grain (P<0.01). PBCR value was unaffected (P>0.50). In conclusion, grain processing increased chewing time and affected chewing regularity in horses perhaps due to decreased palatability.