On Monday the seven-day average of reported U.S. COVID cases sat just below 109,000—more than double the average from two months ago, and up about 10,000 from the week prior. So far, cases in the sixth U.S. wave have largely been fueled by Omicron variants BA.2.12.1 and BA.2.

But levels of BA.4 and BA.5—two relatively new Omicron subvariants thought to be the most transmissible so far, and with the ability to evade immunity from both vaccines and prior infections—are ticking up in the U.S.

Last week the U.S. Centers for Disease Control reported BA.4 and BA.5 in a category with other subvariants because their levels were each under 1%, a CDC spokesperson told Fortune.

This week, however, the agency broke out the subvariants into their own categories. As of last week, BA.4 made up 5% of sequenced reported cases, and BA.5 made up almost 8%, the CDC reported this week.

The question now becomes how well they can compete with dominant variants like BA.2.12.1 and BA.2, which comprised 62% and 25% of infections last week, respectively.

“Even if they can’t compete against BA.2.12.1, they may still increase in frequency here,” Nathan Grubaugh, an associate professor of epidemiology at the Yale School of Public Health, told Fortune.

But what Americans are truly worried about is not which variant wins out, but whether there will be another wave, he said. He sees the likely short-term scenario not as a new BA.4 and BA.5 wave on top of the current wave, but an “extended tail” of it.

Dr. Jake Lemieux, an infectious disease specialist at Massachusetts General and a Harvard Medical School instructor, agreed with Grubaugh, saying Tuesday at a virtual Harvard press availability that BA.4 and BA.5 are likely to be “the tail of the wave.”

Through the July 4 holiday Lemieux expects “fairly sustained transmission.” Beyond that, it’s hard to say.

One thing is certain, Grubaugh said: “It’s likely that BA.4 and BA.5 are not the last of it,”

“What I would put my career on is whether it’s BA.6 or something completely different, there’s going to be another one,” he said.

The White House has predicted a fall and winter wave that could more than double the number of infections the U.S. has seen thus far in the pandemic, as well as a sizable wave of deaths.

What strain would cause that prediction to come true, if it does, is currently unknown. Until recently experts thought that waves were fueled by variants, like Delta and Omicron. But Omicron subvariants like BA.4 and BA.5, which recently swept South Africa, proved that subvariants can cause waves of their own.

“Omicron was so much more transmissible than all the other versions that it would take something quite special to replace it,” Grubaugh said. “We’ve seen within Omicron a lot of ability to continue to diversify and find new niches.”

The question, Grubaugh said, is “if Omicron completely resets the entire pandemic” and its subvariants take over, or if, as experts theorized not too long ago, a new COVID variant, likely to be named Pi, evolves and takes hold.

BA.4 and BA.5 were first detected in the U.S. in late March, as Fortune previously reported. The variants, first discovered in South Africa, swept the country in April and May despite the fact that almost all South Africans have been vaccinated or had COVID.

A recent study out of South Africa found that those who had been previously infected with Omicron but not vaccinated experienced a nearly eightfold drop in neutralizing antibodies when exposed to BA.4 and BA.5. Those who had been vaccinated and previously infected with Omicron saw a milder threefold decrease. 

Alex Sigal, a professor at the Africa Health Research Institute in South Africa, told Fortune in May that symptoms of the new subvariants were similar to typical Omicron symptoms, which include fever, loss of smell, and malaise. 

“I haven’t seen early symptoms of respiratory distress, the major COVID-specific symptom that makes this disease so dangerous,” he said. “It doesn’t feel nice, but there’s less chance of dying.”