There are many ways to split a hive but in any case, the expected result remains the same: A new colony that will raise or adopt a new queen. This requeening process cannot always be guaranteed and may sometimes fail. Precision beekeeping makes it possible to monitor this process from the colony creation. You can therefore identify the colonies that are successful and those that are struggling or even failing without any intrusion.

When to split a hive?

You can make a split without endangering the hive when the brood level is high enough to take 2 or 3 frames. In practice, this stage of “maturity” depends on two major parameters:

1/ The overall dynamics of the apiary, initially characterized by its situation. We can clearly see that the development of the four apiaries below in Paris, Marseille, Pau and Bayonne (France) is very different. The impact of climatic zones is obvious, but the location of the apiary also determines its development. Let’s take as an example the apiaries of Pau and Bayonne which are not very far from each other (114km / 70 miles), but which are displayed very differently.

This year, the French South-West cities (Pau & Bayonne) made a good start. With a mild winter, the egg-laying process started early in January. Marseille had a 15-day time lag, but also showed a very strong growth in just a few weeks. As for Paris, the cooler temperatures explained a slower start, but once this period is over, we can see a very good start!

Average brood volume per apiary
Average brood of four apiaries in France, on different climatic zones

2/ The hive dynamics are also a key factor. Indeed, in the same apiary the colonies may not develop in the same way. This is the case for example in Pau (see graph below) where the first hive started to develop about one month before the last one.

Given these two factors, each apiary and each hive will reach a suitable stage for splitting at a particular time. Whereas in classical beekeeping, the decision is made ” roughly “, in precision beekeeping, it is possible to have a finer indication of the development stage.

I generally plan to split a hive when the colony has reached a 65% brood level. This simple rule helps me to organize the operations in my apiary.

Making splits in practice

Let’s take the apiary in Pau as an example. I decided to split the R3 & R7 hives on March 15. These two hives were the strongest in the apiary and had rocketed the previous days, reaching a brood level of 70%.

Brood volume in the apiary of Pau
Colony development since January 1st. The R3 & R7 hives are ready to be split on 15 March. The other hives are slightly less mature so they will be split later.

To split this hive, I took 2 frames (see the several splitting techniques at the bottom of this article) which resulted in two new hives: RA & RC. No queen was moved or introduced. Therefore, the bees had to raise their new queen from the fresh brood. Under these conditions, the theoretical schedule is as follows:

March 15DHive split with 2 brood frames
March 31D+16New Virgin Queen emerges (at the latest)
April 5D+21All previous generation worker bees have emerged
April 7D+23Queen mating flights start
April 8D+24All previous generation drones have emerged. Colony is now broodless
April 10D+26Egg-laying starts again 3 days after mating

The queen is born on D+16, followed by the period of mating. So, this brings us to D+26 approximately for the resumption of egg-laying (from April 10🤞).

My experience: New queens start to lay between 14 and 20 days after they emerged. Exact day is weather dependent. This would put the start date at around April 14th – April 20th.

Monitoring the colonies’ requeening process after making splits without intrusion

Each of the two new colonies was equipped with a temperature sensor in order to monitor their development. We could clearly see their progress. A few days after the hives were established, we noticed a progressive brood reduction until about April 8. However, from April 10, the two hives showed very different behaviors.

Brood volume in the apiary of Pau
Actual paths of the RA and RC hives stacked with the theoretical calendar. The egg-laying process is effective for RA on the scheduled date, while RC is still buzzing.

The RA hive has developed very smoothly. We can clearly notice the reduction of brood volume during the first phase. From April 10, with the new queen, the egg-laying process starts again. A perfect journey!

For RC, however, things have been more complicated. The first period was identical to RA’s one (both colonies started with 2 brood frames). But RC didn’t manage to recover. The brood level stayed around 20%. Queen was not found during inspection and the presence of scattered drone brood also indicated that the requeening process had failed.

Avoiding brood overheating

When looking more closely at the reasons why the second division (RC) may have failed, I noticed a few overheating alerts.

The measurements below seem to show that overheating (>38°C) affects small colonies. While in the same apiary, larger hives did not suffer from it.

Internal temperature max / Events
Four episodes of overheating in the RC hive impacted brood development.

The RC hive clearly experienced four overheating episodes in April, while RA did not. These heat stresses must have affected its brood to some extent. The lesson learned is that we must be careful of hive overheatings to ensure that splits have the best chance to succeed.

A possible reason for this is that the RC hive is of a more basic construction: thinner walls, no feeder and a basic sheet metal roof, directly placed on the frame cover. As I noticed the overheatings, I decided to add a feeder and a 4 cm insulation under the roof, on April 19. But maybe it was already too late…

The RC hive (on the left) is of a more basic construction compared to the RA (on the right). This makes it more sensitive to overheatings.

The RA hive also experienced overheatings during May. But it was already able to overcome the difficulty. However, its brood development may have been affected. At that time, it still had a frame to build on, although the hive had started on a very good basis.

Did I take the queen when making splits? 😅

Before ending this article, here are two new splits. This time, in a Parisian apiary. They are the result of a 4-frame sampling from 4 different hives, installed in two Dadant hives of 10 frames. The RF and RI hives splitting did not start on March 15 as in Pau, but on April 15.

In this case, we can see that the brood volume drops for the RF hive but not for the RI hive. What’s the reason for this? A parent hive queen may have been removed with the frame. The RI colony developed but the parent hive had to restart. No big deal, but you better know about it!

Brood volume in the apiary of Pau
Two splits that don’t have the same chance of success. One with laying queen (RI) starts quickly while the other (RF) has to raise a queen. Sometime later, in June, they are neck and neck again!

In conclusion: Improving your practice to improve your efficiency

We know how precious a bee colony is today. These examples showed that it is possible to support the development of a new colony without disturbing it. We learned some simple lessons that will help to improve our practice for future hive splits.

The goal is simple: making the maximum splits successful!


References on making splits