1. Events are located in a B-series, only if time exists. In order to see that McTaggart's first premise is correct, one must remember that it is not time in the Newtonian sense—an array of thinglike instants—whose reality is in question. Rather, the consequent of (1)—time exists—is supposed to be construed in a very broad way, as something like "the world exhibits temporality'. And in that case, premise 1 becomes a trivial truth. [Paul Howrich, Asymmetries in Time: Problems in the Philosophy of Science (Cambridge, MA: MIT Press, 1987), 18]
The puzzling thing about the philosophy of time to me is why it is thought that a B-series is ontologically necessary. The idea of "earlier than," "later than" does not need a real B series to function, but rather they could be seen in a B series that is the projection of the movement of an A series. In other words, rather than looking at events according to "earlier than" or "later than," shouldn't it rather be looking at two events on the A series line, project them onto a B series, and only then can it be seen which is earlier and which is later? In other words, events are not located in a B-series, but projected onto a B-series.
Along the same path of understanding, the time words of "now," "past," "present," "future" should be understood as descriptive words not properties of any instance in time. It should be false to describe moment A as having the property "present" or "past" or "future," since whatever descriptor is applicable depends on the relation of moment A to the current "now." Since any one moment does not possess temporal properties of any sort, but rather everything is merely relation, the whole idea of any one moment A acquiring simultaneously the "properties" "past," "present," "future" as the "moving now" passes through that moment is without merit.